Human Players Should Have a Higher Valued Like/Dislike Vote

Greetings to all,

As a previously made post outlined, Roblox’s like/dislike voting system is a little clunky and bypassable. One problem that is widely known is the mass dislike botting where thousands of bots join a game for not even a second, then they dislike that game to sabotage its ratings. Although smart developers and the Roblox staff have taken measures (i.e. through modules), the problem is still existent. But, I believe that I have found an antidote for that.


Problem Overview

As a Roblox developer, I (among others) feel very insecure in publishing a game because I know, some time or another, my game is going to be the victim of dislike botting. Ratings determine the quality of a game, and it is unfair when a bunch of things run by computer code come and sink the reputation of a quality game.

Potential Solution

But I do think that there may be an easy fix for this. First, Roblox can create an algorithm to verify that players are human. Perhaps, Roblox can automatically mark all DevForum members and new members verified because they successfully passed its entrance requirements (which has no flaws for the most part). Anyways, after that, any of a verified player’s like/dislike votes can be amplified by a certain factor (greater than before) because that person is genuinely human. This is why this solution has a strong potential: every bot is a newly made account that is not even close to being verified (depending on the algorithm that Roblox would set) and thus, their unfair dislike on a game would be no match to the real votes by real people. This would stabilize and secure a game’s ratings prevent it from being damaged too much from any attack such as these.

For example, while a non-verified players’ votes would be 1 each, all verified players’ votes can be a 5 each (or just any reasonable number higher than 1).

Effect of the Solution

If this feature is added, then it would not only ensure that all games’ ratings are genuine and not unfairly influenced, but it will also mean that more developers would be encouraged to create a game. They would know that their game is safe and secure.


TL; DR

To combat against mass dislike botting, Roblox can simply value verified-as-human players’ votes more to stabilize and bring games’ ratings to its true number.


I understand that this is more like a temporary fix, but it will handle the majority of this bot issue. We need to take precautionary measures first, then work on completely eradicating the problem.

I hope you support this idea,
thank you for reading!

8 Likes

This is a miniscule amount of people compared to Roblox’s regular user base. It would leave no impact whatsoever unfortunately.

There is no fool-proof way to determine if a user is a human or not (Roblox would already be applying this across the platform to ensure bots cannot log in at all, if such a check is known). Bot accounts have a verified email and whatever else it will take to get past the mechanisms Roblox has put in place to prevent botting.

7 Likes

That is just an example for Roblox to start the verification process off with. There are additional criteria that Roblox can set up to filter out humans and bots such as account age, interactions with other players, created games, and in-game/website purchases (which are all effective).

Yes, this is quite understandable. But, wouldn’t the aforementioned interactions with players, created games/places, and in-game/website purchases succeed in keeping identifying bots? First, chats with other players, whether that is online or in-game, can portray the user’s intentions and humanness. second, if a user has created some sort of game (that is different from the preset templates) and has a decent amount of visits and favorites, then it is evident that only a human can accomplish such a thing. Last, but not least, if a user has purchased anything via Robux or real-life currency (membership, catalog items, developer products, gamepasses, etc.), then it definitely means that they are human.

Considering these several ways of determining humans and bots, are they not efficient or effective enough to combat dislike botting to a reasonable extent?

2 Likes

That wouldn’t actually be a particularly good way of dealing with bots, since if a bot can detect it can’t log in, then it’s pretty obvious they’re blocked and need to work on a new way of getting around the checks. Ideally, they’d make it look like the bot is logged in and performing actions successfully (from the perspective of the bot) but silently discard their actions for everyone else…

…but this would be better suited to it’s own thread. Just a few thoughts.

2 Likes

Would verifying human players using these ways work?

  • Premium subscription
  • Some Robux in their account
  • Verified Email address or phone number
  • Veteran badges

Credits to @monomorphosis (via PM) for the first , third, and fourth ideas.