What happens when you make a move in lichess.org?(davidreis.me) |
What happens when you make a move in lichess.org?(davidreis.me) |
Regardless, one thing I find maddening about chess.com is the time architecture of the game. I haven't seen the underlying code, but it feels like the SERVER is tracking the time. This completely neglects transport time & latency meaning that 1s to move isn't really a second. Playing on the mobile client is an exercise in frustration if you are playing timed games and down to the wire. Even when you aren't, your clock will jump on normal moves and it is most obvious during the opening.
This could also be due to general poor network code as well. The number of errors I get during puzzles is also frustrating. Do they really not retry a send automatically?? <breath>
Chess.com has the brand and the names... but dang, the tech feels SO rough to me.
This is one of the many, many things but imo it's the most telling. They can't even add a clock counting down the 6 minutes to their web client.
TBH this is what I expected for all online chess. How else to reconcile the two players' differing clocks and also prevent client-side cheating?
I do recognize that fps games utilize predictive algorithms and planning to estimate future player positions but still, turn based networking with 100ms accuracy should be a solved problem
Is there a point in preventing cheating, really? I can just make a bot...
Seems right.
If you export/download games from lichess, they use the .pgn (Portable Game Notation) format, which is a standard plain-text format circa 1993, used by pretty much everyone for describing a chess game.
Lichess follows the specification to the letter, and as it only technically allows one-second accuracy, lichess only record moves with one-second accuracy. It seems insane, but that's how they do it.
Chess.com also exports PGN files, but they add a decimal place, allowing subsecond accuracy. No one has a problem with this. There is no software which cannot handle this. But Lichess refuses to "break" the spec.
lichess PGN export example:
> 1. d3 { [%eval -0.15] [%clk 0:01:00] } 1... g6 { [%eval 0.04] [%clk 0:01:00] }
Chess.com PGN export example:
> 1. d4 {[%clk 0:02:58.6]} 1... b6 {[%clk 0:02:59.2]}
According to this blog post, this doesn't appear to be the case since at least 2017:
https://lichess.org/@/lichess/blog/a-better-game-clock-histo...
"Move times are now stored and displayed with a precision of one tenth of a second. The precision even goes up to one hundredth of a second, for positions where the player had less than 10 seconds left on their clock."
It's super annoying and the reason I only play blitz+ on chesscom.
[1]https://www.chess.com/forum/view/help-support/mate-in-one-qu...
chess.com confirmed the issue.
Interesting that they accumulate and periodically store game state. Unfortunately it is not very clear, where they store ongoing game state - in redis or on server itself. Also cost breakdown doesn't have server for redis, only for DB.
BTW, their github has better architectural picture, than overly simplified one in the article: https://raw.githubusercontent.com/lichess-org/lila/master/pu.... Unfortunately, I'm afraid, drawing something like that during interview may not land a job at faang =(
Note that they have cost per game fairly low: $0.00027, 3,671 games per dollar.
Their cost breakdown, for ones who are curious https://docs.google.com/spreadsheets/d/1Si3PMUJGR9KrpE5lngSk...
p.s. I'm not saying that Lichess's approach is the best or faang is the worst. Remember, lichess had 10 hours outage exactly because of the architecture chosen (single datacenter dependency). https://lichess.org/@/Lichess/blog/post-mortem-of-our-longes... . And outages like that are exactly the reasons why multi-datacenter and multi-region architectures are drilled down into faang engineers.
My point is is that there are cases when this approach is legit, but typical interview is laser focused on different things, and most probably won't appreciate the "old style" approach to the problem. I'm sure that if Thibault will ever decide to land in faang he will neither do whiteboard coding nor system design.
Just a wild guess: might be intended to lower the implementation barrier for new open-source software clients on new platforms, and/or preempt them from implementing subtle logic bugs that only show up much later.
The rules of chess are a bit tedious to implement, and you can easily get tired and code an edge-case bug that's almost invisible. Lichess itself did this—it once had a logic error that affected a very tiny number (exactly 7) of games,
https://github.com/lichess-org/database/issues/23 ("Before 2015: Some games with illegal moves were recorded")
(I apologize I couldn't find the specific patch that fixed this)
I will have to take a look, because whatever it's doing, it works very well!
Well, except for that one major outage where everything shit the bed due to some misconfiguration of IP multicast in the datacenters, or so I was told.
So, maybe if your mission isn't life critical, you can just wrongfully assume exactly-once delivery.
[1]: https://en.wikipedia.org/wiki/Pragmatic_General_Multicast
To do that, the server needs some measure of “how long does the client think the player actually took to make a move”, to later subtract latency not attributable to actual thinking from the clock.
I tried this and not all the messages I sent arrived.
Saw CF had some paying solution, but was wondering about a free solution
Cloudflare has a lower latency product called Argo Smart Routing [1]. When we tried Argo in 2020, we still saw 10+ ms increased latency across the board, which is unacceptable for competitive multiplayer games. That said, Discord voice still (or used to) uses Argo for voice, so there are certainly less latency-sensitive games where it would work well.
The other issue with sockets over Cloudflare (circa 2020 on business plan) is they get terminate liberally with the assumption you have a reconnection mechanism in place. I'd imagine this is acceptable for traditional WebSocket use cases, but not for games.
Services like OVH & Vultr also advertise "DDoS protection for games," but I've found these to be pretty useless in practice. We can only measure traffic that reaches our game servers, so I have no way of knowing if they're actually helping at all.
Your best bet is getting familiar with iptables and fine-tuning rules to match your game's traffic patterns. Thankfully, LLMs are pretty good at generating these rules for you nowadays if you're not already familiar with these tools. Make sure to set up something like node-exporter to be able to monitor attacks and understand where things go wrong. There have been a few other posts on HN in the past that go into more depth about game server DoS mitigation [2] [3].
I built something in the same vein for my startup (Apache 2.0 OSS, steal our code!) [4] that runs a series of load balancers in front of game servers in order to act like a mini-Cloudflare. In addition to the basics I already listed, we also have logic under the hood that (a) dynamically routes traffic to load balancers and (b) autoscales hardware based on traffic in order to absorb attacks. We're rolling out a dynamic bot attack & mitigation mechanism soon to handle more complex patterns.
[1] https://www.cloudflare.com/application-services/products/arg...
[2] https://news.ycombinator.com/item?id=35771466
What still isn't great is the ecosystem and the build-tooling compared to Rust (part of it because of the JVM). But just language-wise, it basically has all the goodies of Rust and much more. Ofc. it's easier for Scala to have that because it does not have to balance against zero-overhead abstraction like Rust does.
Still, Scala was hyped at some point (and I find it wasn't justified). But now, the language is actually one if not the best of very-high-level-languages that is used in production and not just academic. It's kind of sad to see, that it does not receive more traction, but it does not have the marketing budget of, say, golang.
> all the goodies of Rust
Does it prevent me from using a non-thread-safe object in multiple threads? Or storing a given object which is no longer valid after the call ends?
Does it have a unified error handling culture? In Scala some prefer exceptions (with or without `using CanThrow`), some prefer the `Either` (`Result`) type.
Does it have named destructuring?
I don't understand why the author didn't just look this up in the source code. Lichess is open source and we can see exactly what this field is here, it's the average lag:
https://github.com/lichess-org/lila/blob/45b5f0cfbbf6c045ad7...
send = (t: string, d: any, o: any = {}, noRetry = false): void => {
const msg: Partial<MsgOut> = { t };
if (d !== undefined) {
if (o.withLag) d.l = Math.round(this.averageLag);
if (o.millis >= 0) d.s = Math.round(o.millis * 0.1).toString(36);
msg.d = d;
}
if (o.ackable) {
msg.d = msg.d || {}; // can't ack message without data
this.ackable.register(t, msg.d); // adds d.a, the ack ID we expect to get back
}
const message = JSON.stringify(msg);
...
private schedulePing = (delay: number): void => {
clearTimeout(this.pingSchedule);
this.pingSchedule = setTimeout(this.pingNow, delay);
};
private pingNow = (): void => {
clearTimeout(this.pingSchedule);
clearTimeout(this.connectSchedule);
const pingData =
this.options.isAuth && this.pongCount % 10 == 2
? JSON.stringify({
t: 'p',
l: Math.round(0.1 * this.averageLag),
})
: 'null';
try {
this.ws!.send(pingData);
this.lastPingTime = performance.now();
} catch (e) {
this.debug(e, true);
}
this.scheduleConnect();
};
private computePingDelay = (): number => this.options.pingDelay + (this.options.idle ? 1000 : 0);
private pong = (): void => {
clearTimeout(this.connectSchedule);
this.schedulePing(this.computePingDelay());
const currentLag = Math.min(performance.now() - this.lastPingTime, 10000);
this.pongCount++;
// Average first 4 pings, then switch to decaying average.
const mix = this.pongCount > 4 ? 0.1 : 1 / this.pongCount;
this.averageLag += mix * (currentLag - this.averageLag);
pubsub.emit('socket.lag', this.averageLag);
this.updateStats(currentLag);
};Edit: also includes move count but not repetition.
https://en.wikipedia.org/wiki/Forsyth%E2%80%93Edwards_Notati...
Even though nowadays I hardly have time to play, I'm still happy to support such a delightfully honorable and usable(!) open-source project.
It's a weird trend. Altruism truly does not exist
(I donated btw) (Probably more than you) (But who's counting)
There are many aspects in which they are not the best.
Ad-free, compute intensive, non-CRUD, massively scaled, complex cheat moderation, infinite puzzles/analysis, educational (studies/tactics/openings explorer), etc. All this for free. I'm curious what's the best website in your opinion
Yet another reason to be skeptical of the quality of hiring in faang if anything.
Naturally, it's not possible to view this move anymore, but this game (https://lichess.org/XDQeUk6j#48) has everything up until the last legal move right before the illegal castling happened.
Also that linked game is pretty entertaining. It's not a good game, but it can be fun watching lower ranked players make moves that you'd never see in higher level games. Like, who plays Bb5+ against the Scandinavian? Amazing stuff.
(the broken code checked that the only pieces on the king's path to its new position were kings and rooks of the appropriate color)
Validating a submitted move is distinct from listing valid moves. I assumed the server would need to validate regardless of providing a list to the client.
A bit of surprise consideration … is that even common in these days of overfancy web sites.
>>I think I heard a streamer say it was for kicking out the cheaters
I can't see how it can possibly help. Maybe he meant something else?
on lichess it does have an impact. lichess has a thing they call lag compensation where the server can add time to a player's clock after the server receives their move.
The goal is to make it fair for someone with high lag playing someone with low lag.
I don't know the exact cheating method used. I'll have a guess, though. What if someone spent a few seconds looking at the board before making their move, and then adding (edit: oops, subtracting) a few seconds to their clock in their response packet. The server would see the client made their move instantly based on the time in the response packet, but it took a few seconds for the server to receive the packet. i.e. lag. So it might add time to compensate for the perceived lag.
Lag compensation cheating is a frequent topic on the lichess forums.
My criticism was mostly towards the very poor metrics these companies have introduced behind hiring, albeit I can understand that given the gigantic amount of applications they get a mechanism for removing false positives is acceptable even if missing on false negatives.
And even more that it spread to companies that do not have their problems and can't afford false negatives.
Looking at second-order effects, many companies look up to FAANG for "best practices", which often includes them blindly copying their hiring practices. Without feeling or calling out any healthy skepticism, the software hiring world becomes a worse place overall.
At any rate my conclusion was disappointment that if I actually want reliability, I need to implement my own ACKs anyway, meaning I'm paying a pretty high overhead for no benefit.
At least now there's UDP in browser with WebTransport. I haven't tried it yet, but I hear it's a lot more pleasant than the previous option WebRTC, which was so convoluted (for the "I just want a UDP socket" usecase) that very few people used it.
And 2.) most people will go with sbt; and while it has improved a lot it is still comparably slow, has some annoying bugs and so on.
Compare that to Rust - I don't think those problems exist there.
Basically, you needed a good and experienced developer from the start of a project for it to be a nice code base.
> I'm very fluent in Scala 2, but I will avoid Scala if I can, mostly to stay away from purely functional programmers.
There is the whole [Li Haoyi](http://www.lihaoyi.com/) ecosystem in Scala that is much more python-like, but nicely designed, statically typed and using immutable datastructures by default. I think it's the best you can get nowadays if you want to have immutable datastructures on the JVM. Any other option I've ever tried was way worse.
If you are fine with Java's stdlib then I guess Kotlin is the better choice.
> Does it prevent me from using a non-thread-safe object in multiple threads?
I would answer the question with yes, but maybe in a different way than you might expect. Scala prevents problems/bugs from using a non-thread-safe object in multiple threads by simply having immutability by default. Rust cannot do that (due to performance) so it has to have another way (the borrow checker). I would argue that the Scala way is better if you don't need the performance / memory-efficiency of rust and can live with garbage collection. That reduces the domains that you can use Scala for, but in exchange the code will be simpler compared to Rust code, so in those domains Scala will have the advantage but it's a minor one.
> Or storing a given object which is no longer valid after the call ends?
To this one I would say "in practice yes". Rust is better here, but when using e.g. [ZIO Scope](https://zio.dev/reference/resource/scope/) then the problem isn't really existing. You can technically still do something like that, but you would basically have to do it intentionally. Rust has the advantage here though, but it's a minor one.
> Does it have a unified error handling culture?
No, Scala has no unified culture. Maybe the situation is better than in Rust, but then Rust has its own problems. [Just a few days ago I found a comment about a problem caused by a hardcoded panic that caused issues](https://github.com/orgs/meilisearch/discussions/532#discussi...).
> Does it have named destructuring?
Unless we are talking about two different things, yes it does. I would even argue that Scala is more powerful here, because it also supports local imports and (with Scala 3) exports. So not only can you extract fields of an object into a variable, you can also generally bring them into scope and alias them at the same time, but you can do the reverse as well: [you can export them as well](https://docs.scala-lang.org/scala3/reference/other-new-featu...).
He has very good taste. I wish more Scala people are like him.
> simply having immutability by default
Not everything is a pure data structure. I called a gRPC streaming callback with multiple threads in Scala (got garbled result in the receiver). You can say this is the fault of using the Java API, but the more Scala solution (fs2) involves serializing the access under the hood which is not cheap.
Recalling that my contention is with "all the goodies of Rust".
>> named destructuring
> Unless we are talking about two different things, yes it does.
I guess we are. I complained about this quite some time ago: https://news.ycombinator.com/item?id=31399737
Well yes, that's what I'm saying: in Scala you sometimes have to sacrifice performance. Though I don't think that serialization is generally required just because you use e.g. fs2 or ZIO for streaming.
> I guess we are. I complained about this quite some time ago: https://news.ycombinator.com/item?id=31399737
You are making a fair point. I'm also not happy with some of the decisions about pattern-matching.
You can resolve some of those issue by just using `import` locally though. So for example, if you have a case class X with many fields and you want to access many of them, you don't need to extract them all in pattern matching (and deal with _ for the stuff you don't need) but you can rather do `val myX = X(...); import myX._` and then just use the fields.
That is basically equivalent to doing `const {a, b, c} = myX` in typescript. You don't need to do `case X(a, b, c, _, _, ...)` in Scala here.
Not saying that this just solves all issues, but I think ergonomics of Scala in terms of pattern matching, destructuring and scoping/importing is generally/overall not worse than in Rust, at least not significantly so - that's why I think it's fair to say that Scala has the same "goodies" in this area. I did not mean to say that Scala is as good as or superior than Rust in all language features.
Carlsen, Nakamura and chess.com itself have participated in the Niemann witch hunt. No evidence of over-the-board cheating has ever been provided.
None of the accusers is banned.
Kramnik did the same as they did (by indeed going too far).
Yes the instance of chess is finite but the problem of computing moves is inherently in NP.
The key is that just because a problem is in NP it does't mean that its difficult to solve the instances with small parameters.
See the famous coloring, SAT, or any other equal NP problem...
You can read about what became timeseal here. https://eprint.iacr.org/2004/203.pdf
Yes, it's easy to cheat with this, but it's very easy to cheat with chess anyway.
What is the point of responding with any legitimate criticism when any potentially negative sentiment however mild, upfront, expressing disagreement, gets downvoted to the point where the mechanics of the website squelches the person and silences them (by purposeful intent).
Can you ever have any legitimate intelligent conversation after a participant has been harmed and effectively silenced in this way?
When you cannot speak freely, there can be no intelligent communications raising the bar objectively. The opposite occurs, and anything provided, even seemingly rational conversation falls after such a threat or action of violence, all conversation then falls into the gutter as a result of the added coercive cost imposed. You may contend that its not violence, but it meets the WHO definition for such which properly accounts for psychological torture and coercion (of which this is a common form).
It should go without saying, but you cannot have any intelligent conversation when those who embrace totalitarian methods prevent you from speaking (and yes these meet the criteria).
At the point this happens, regardless of valid criticism, or pointing out errors in methodology, it all dies on the vine, the communication is clear; you will be punished for disagreeing. That destructive behavior inevitably leads to ruin.
This is fairly basic stuff, in order to think and be intelligent, one must be able to risk being offensive. In order to learn something new, one must risk being offended.
When neither are possible because you or someone else muzzles any conversation expressing disagreement or corrosively add cost, even under such modest terms as here, the fallout is silent, yet devastating.
It might not seem like much, but the light goes out of the world as those with intelligence withdraw their support, and the natural consequences which were held at bay by these people, albeit slow moving, become inevitable.
Best of luck to you. There is only the possibility of harm by continuing any discussion under these circumstances.
I'd suggest remembering this when you start wondering, "where have all the intelligent and competent people gone?".
Silence doesn't indicate agreement. It is indicative of the best and brightest no longer contributing to the same systems that seek to destroy or enslave them.
But I believe it was just that: a negative sentiment. Not exactly a "constructive, intelligent criticism". And when you go there, the reality is that people will vote to reflect their own opinion. If you say "This project is so amazing!" and get a ton of upvotes, it does not mean that your comment is super useful; just that many people agree. Similarly, if you say "Naah, it sucks" and get a ton of downvotes, it means that many people disagree. Not that they want to silent you.
Now try an actual constructive criticism: you may get downvotes (that's how it is because people are emotional beings), but probably upvotes as well if you bring interesting insights.
> There is only the possibility of harm by continuing any discussion under these circumstances.
That's fair. I think one mistake there is that you should have started with a constructive criticism rather than an admittedly polite "naaaah, I think it sucks".
You might suggest such, but this has the effect of just baiting me for a response so it can be marked down more where you are engaging me for the effect to further punish.
You see these people don't do this because of their opinion, they do it because it causes psychological harm, its a totalitarian tactic that is not unknown. Silencing was used somewhat heavily during Hitler's rise to power.
Forcing the only conversation to first agree before moving forward, at any point, causes you to fight your own psychology to remain consistent and the process warps you subtly. Most aren't self-reflective enough to notice but the effect is the same regardless.
Robert Cialdini wrote quite a lot about the various lever of influences that are often used as mental compulsion/coercion, and Joost Meerloo and Robert Lifton both cover these structures and techniques in detail in the context of WW2 torture and moving forward. These behavioral structures run parallel with those of the Nazi's, and other totalitarian regimes.
This is what is happening, and when mild conversation causes this type of behavior, this is the time you should be most greatly concerned because its arbitrary, causes mass delusion, and continues until destruction, albeit slow, overtakes that group.
As far as I'm concerned, the people doing this can ride their train right to their own demise for all I care. They are true evil, and they'll be doing the world a favor when that happens. The rules of society will no longer protect them once they destroy society.
We were taught from children to not be violent. This is violence, there is no excuse for bullying, and people are no better than animals if they can't reason and be civil. What one does in small things, they do first in large things that matter.
If they want to be violent for a mild comment like that, they won't get anything from me, and I'll reciprocate in the only way I can right now, withdrawing and not providing anything of value.
I'll pray I never meet them in person because if you or anyone else tries to harm me, I'll be exacting an equal or greater cost in self-defense.
This destructive behavior is despicable on so many levels, and you say its not so bad but you don't realize just how bad it gets, this behavior promoting menticide, and robotization is what led to the gas chambers in Germany during WW2.
When no one questions rationally, or can express disagreement, evil flourishes. You can't ever argue with evil, you have to kill it, as we had to do during WW2 (at great cost).
Read the notes from the Wannasee Conference, or if you can't be bothered, rent Conspiracy (2001). The history is well documented by experts who studied these things to prevent it from ever happening again, and yet it seems no one has learned their lessons since they repeat it yet again.
They are emotional beings (/s), can you imagine that being a valid defense of Nazism during WW2? For the deaths of all those Jews in the camps? If it's unjustifiable at the extremes, it is unjustifiable anywhere.
These are the same things, the only difference is perspective and the fact that you don't have perfect information upfront at the bottom level, you only ever find out afterwards, and its a goose-step death march ever forward and people don't realize this is how it works. One step at a time, pivoting, with no questions.
This is only the beginning, and when you can't stop it early, then its too late to do anything later to prevent the massive destruction that these people inevitably bring on themselves and everyone else.
This is why it is so damn important to protect and maintain freedom of speech in a civil atmosphere. There can be no rational support for this behavior, not ever.
Please stop making excuses for the truly evil.
Redundancy, scalability, decoupling, resilience, best possible handling of errors, cost optimization, etc. may be more important at the scale Netflix operates at.
So much that they built a tool to intentionally make things difficult (read: it arbitrarily stops production system processes/containers/etc.) and help inform what decisions to make in favor of fault tolerance.
> Exposing engineers to failures more frequently incentivizes them to build resilient services.
TCP, although I like to imagine FPS games where shooting someone sends "DELETE /players/n00b HTTP/1.1" to the server.
It is actually perfectly fine to not ban people doing something wrong one or two times while banning people doing that exact thing for fifth time.
The former is always excused, presumably because people's ancient group instincts kick in. The latter is a single heretic that must be destroyed.
The group does much more damage than the individual. Everyone already thought that Kramnik's pseudo-mathematical evidence was garbage, so why the ban? The answer is that he mentioned well connected people like Nakamura, so the rogue nail needed to be hammered in.
Groups are never punished, and 95% of Internet commenters always excuse the group.
Niemann was already an admitted cheater who had been previously sanctioned for his activities. Carlsen’s accusations against him may have been unfounded for the specific game in question but Niemann’s reputation was already blemished. Niemann’s lawsuit was inexcusable though, as it was essentially a SLAPP [1] designed to hinder everyone’s efforts to get cheating out of chess. Thankfully the lawsuit was unsuccessful and Niemann continues to play chess.
[1] https://en.wikipedia.org/wiki/Strategic_lawsuit_against_publ...
Now I would not compare this to Nazism or call it "true evil". Try to pick someone randomly in the street, go talk to them and politely explain why you find them unattractive (e.g. "I just would like to say that anyone finding you attractive would have pretty low standards"). Would you call it Nazism if they asked you to leave them alone?
Isolation does weird things to the mind, one of which distorts reflected appraisal, other parts where reflected appraisal is denied in communications are even more impactful. It induces a involuntary hypnotic states which varies in intensity by exposure.
The tortured takes on mannerisms of the torturer, and when denied communication, or only distorted reflected apprsail long enough their entire being unravels, and you have psychotic break or disassociate completely. The psychotic break is a semi-lucid state where planning is capable. To make an apt comparison, an active shooter might fall into this latter category. Years of investigations into what causes these people to do what they did, and the powers that be can only say with certainty that these people were bullied, but no clear cause can be found.
Coercion is a dangerous thing.
You can corroborate what I've said with the literature in the books I mentioned, or with isolation studies where the studies had to be cancelled early for the safety of the study participants. There is a lot of research out there.
Destroying someone's identity, their personality, what makes them them, is true evil, and these structures are how you do it which is why its so important to recognize the problem, few do.
Your comparison is apples to oranges. It is not asking someone a opinion, its silencing them entirely by force, where they are disadvantaged when they don't answer. There is a very big distinction between the two.
I do understand, and it is unfortunate that moderation gets mixed up with score. If you have a better solution, feel free to explain it! Because not moderating at all is a problem in its own.
Okay, I can’t keep this up. I was parodying the position not being serious.
You only allow moderators to do the actual moderating. You don't allow upvotes or downvotes because they are used following a sybil attack structure (many sock-puppet accounts to one person) to silence or amplify messages.
You have a flag button (which they already have in place for HN), to report content that should be flagged for violating rules.
Those that report spurious (good) content get warned/punished for abusing the flag features. They may have the feature silently revoked (shadowbanned) for abuses, or banned for other activity suggesting the accounts are sockpuppets (i.e. going directly to an article when normal viewing requires first loading the index, then following a link with the associated metadata including the referrer to get to the page to make a report.
It really is that simple.
This moderates, and it limits bad actors, its still done in most online forums that are still around; because it works.
What about this? Sounds like it may "silence" people who flag stuff that [those in power] think should not be flagged. How is that different from you thinking that you are silenced because people did not like what you wrote? Someone could feel like they get silenced/warned/punished because they genuinely flag what they see as inappropriate content, right?
In practice, when you write clear consistent community rules and guidelines that are unambiguous, the averse silencing issues of arbitrary action are non-existent. The only people who get warned are bad actors, or others with mental illness often based in delusion (who should be de-amplified and silenced as they are unwell).
Echo chambers promote the spread of mass delusion and the psychosis needed by totalitarian regimes to persist in their destructive natures towards a final outcome. This includes state-run media (a tyrant's best friend).
These use the same mechanisms we use naturally to develop our identity and adopt our culture as we grow from children to adults, but towards more destructive outcomes, the banality of evil and the radical evil (WW2) are well studied subjects.
It should go without saying these moderator actions all inherently come with reasonable graduated responses based on severity and persistence of the threat; sophistication naturally increases severity, but remains consistent unlike HN news.
You are told what you did wrong when you do wrong as opposed to the Chinese Anaconda in the Chandelier Strategy whose outcome will only result in all rational people being eaten first, then the irrational rest, then no ones left (since the Anaconda dies from not eating and it ate everyone).
Being told what rule you broke and having it be reasonable, fair and just, is fundamental in having due process, and underlies the basic "rule of law", and avoids conformist irrational silencing by mob or corrupt officials (as under a "rule by law"; you show me the person I'll show you the crime.)
> How is that different from you thinking that you are silenced because people did not like what you wrote.
At its core, you have to ask yourself what is the difference for you between objective reality and delusion, and secondarily (or primarily), what is your working definition of delusion. Finally, should the delusional be allowed to force you to become delusional through mental coercion? Those who are delusional are fundamentally destructive because they are blind to the truth (a requisite of evil people).
Needless to say, this is the bundle of questions underlying the core question you are asking, though it has intrinsic ties to how we know truth from falsity in the first place which is based in western philosophy. Unfortunately, this subject is not taught in developmental curricula anymore (K-12), and classical education covered this under the curricula that included both Trivium and Quadrivium curricula elements pre-Prussian school model of centralized education. The classical education gave rise to generations of hyper-rationalism which allowed rapid technological progress and advance up until WW2 when the change (given maturity time lag), occurred. At any one time two to three generations live side-by-side, a generation being 20 years. Those raised under the prussian model came of age in the 1930s-1940s. The late 1940s-1970s are when societal problems started occurring beginning at the critical point an individuals 30s-40s where political power is transferred generationally. It progressively has gotten worse ever since to current day.
Unfortunately, answering those questions without allowing ambiguity would exceed several pages and laying it out step-by-step so you can see the mental gymnastics done to point out the subtle distinctions, make this area fairly irreducible in a post here (its too long).
At its core, the issue is often started with corruption of language and definitions which involves creating circular or contradictory alternate definitions (not tied externally or objectively). As an example, take a look at the google definition for delusional (2), and then deluded, both claim to be from the Oxford Dictionary (from google). What happens when you say something based in fact, and they adjust the context deceitfully using these contradictory definitions.
Once you check those definitions out, then go to the Oxford Dictionary website and find that definition pre-1970. You will find there is a subtle but important distinction and that they are not the same. Newer similar ambiguous words are being adjusted to be tautological as well. The false premise being if you don't have words to describe something accurately, then it doesn't exist (often an underlying communist/marxist thought theme).
This lessening is what Orwell tried to warn against in his dystopian book related to language, deceit, and its impact on people.
This isn't an accident either, it is being done purposefully by intent and design following socialism/communism themes. The selfish ledger is a leaked google document that uses elements from Maoism in Google's own words, and debunked researchers, to mislead and promise utopia while leading to dystopia. This is why its important to keep older dictionaries when online references can be poisoned/paywalled without notice.
Check out Dr. Epstein on Youtube related to Google's Worst Enemy. You'll no doubt notice he's being discredited algorithmically by matching related videos up with Jeffrey Epstein to manipulate you. This is the common level of sophistication and it goes so much deeper in the literature, and its rationally supported.
To answer your core question.
The delusional or schizophrenic person doesn't base their reality on external objective evidence following rational principles, or structure. When someone does base what they say with evidence/fact, they aren't and can't be delusional until they use something incorrectly that may be valid but is not sound. Baseless feelings don't generally matter, they are transient where objective facts and supported premises do matter (above all else).
> Someone could feel like they get silenced
People can feel whatever they want, guidance is issued by moderators when something is unclear to remain consistent. Unwell people are encouraged not to participate. Using references to words like feelings, and other aspects of the mind all boil down to whether the person in question is being rational or delusional. If the latter, they are unwell and should be moderated. If the former, they can participate. The rules are consistent to segment the two based on rational behavior.
Rationality and our ability to discern falsehoods is largely what separates us from animals, and we are living through an age of lies, deceit, and ruin never before seen. Taken to one of many possible logical conclusions, eventually dependent systems will collapse as all systems have been made progressively more brittle as concentration/sieving occurs at all levels of centralized planning (predicted), and eventually just like under Mao, there will be a great dying only it will be much worse. A calamity, never before seen at such a scale. This is the natural consequence when you have a generation die off who disadvantaged their children through changes they made, so profusely that they don't have a replacement birthrate (i.e. depopulation).
As you can see many things are intertwined but they largely all come back to the fundamental building blocks of society and what made it work previously, which has been changed and made more brittle systematically over time, prior to bringing stress to cause failure for a pivot to statism.
If one isn't educated on what those building blocks in the first place are, its difficult to even have a discussion, or get traction on problems since there is no common ground for shared meaning.
Putting it mildly more damage is done with unsound methods based in delusion, and an incomplete picture on the failure domains, than no action at all. People today largely were never taught to think rationally. A very small minority are an exception, and they have no real voice.
Cascading failures on two sides of a shrinking cliff-side, with the delusional leading the way over the cliff like lemmings, and preventing any other path forward secure in their hubris and death march towards destruction.
It is not a hopeful outlook once you get to a point of no return, but that's how it is when people become complacent and eventually delusional.