Answered Automated Report Bot = lazy people (1 Viewer)

Status
Not open for further replies.

PNN

Trial Gameserver Admin
Staff member
Trial Gameserver Admin
Donator
Joined
Jan 27, 2016
Messages
877
Okei so the title says it all, and I want it gone, aswell as many others.

Before I start with the thread itself I will say this, I don't really know where to place this thread, so I guess questions is the right place.

There is many issues with having this bot, and I will come to them. First off, seriously, how lazy are we when we need a bot to report minor and not so minor things in chatlogs??

So the bot reports people who has 5 strikes in the chatlog and who has 1 comm-block already, which means the entire USJB server can be reported. Say if an admin is on, and I say something that the bot will detect. The admin has warned me, however the bot reports me. This means that the bot is broken and the report is invalid. If the admin used some sort of /csay to warn, and the player who gets reported did not screenshot nor get any other proof of it, the bot will win the report. This will be a problem.

Some of the explenations of how the bot works from my side might be false or not, but my opinion remains the same. It sucks.

My question is to Kevin and everyone else, even though I know the JB players will support the bot, how is this allowed? Why is this a thing?

Looking forward to see some explenations.
 

tres

Well-Known Member
Joined
Feb 10, 2020
Messages
42
There is many issues with having this bot, and I will come to them. First off, seriously, how lazy are we when we need a bot to report minor and not so minor things in chatlogs??
How should discriminatory words such as the n-word, fa-word (one against homosexual people), and words that offend people's sexuality be overlooked?
It is an obvious violation of [2][a],
1599077037066.png

Discrimination isn't tolerated and should not be tolerated in gameservers, so there should be no exceptions to disregard these.

I do not understand what you mean by how it is lazy to report these messages. These players are obviously breaking the rules.
1599076957106.png
So the bot reports people who has 5 strikes in the chatlog and who has 1 comm-block already, which means the entire USJB server can be reported. Say if an admin is on, and I say something that the bot will detect. The admin has warned me, however the bot reports me. This means that the bot is broken and the report is invalid. If the admin used some sort of /csay to warn, and the player who gets reported did not screenshot nor get any other proof of it, the bot will win the report. This will be a problem.
I added a limit of 5, players will mostly stop after 2-3 flagged messages, but if it still continues, then it will be considered a problem.
The limit of 5 only includes players who have 0 commbans. Not 1 or more commban.
I also set a minimum date limit, meaning my bot does not go beyond old chat logs. Admins complained that without a minimum date requirement, then gagging the player would feel "out of context."
 

TRGGB2

Well-Known Member
Joined
Mar 16, 2019
Messages
290
Here's my opinion.

Yes it has flaws but it has a purpose.

I think we can use it on servers on if:
-The server is mostly popular but doesn't/hardly has any admins on it.
-The server has not too many regulars reporting rule-breakers.

I think we should not use it if:
-The server has many active admins doing warnings
-The server has regulars reporting rule-breakers.

Thats my opinion at least. @PNN any counterarguments with this?
 
Last edited:

PNN

Trial Gameserver Admin
Staff member
Trial Gameserver Admin
Donator
Joined
Jan 27, 2016
Messages
877
I do not understand what you mean by how it is lazy to report these messages. These players are obviously breaking the rules.
Let some of the players who actually play at the time handle the report then. It's lazy.

As for the limit of flagged messages, you don't know if the player was warned or not. Period.

I think we can use it on servers on if:
-The server is mostly popular but doesn't/hardly has any admins on it.
-The server has not too many regulars reporting rule-breakers.
There's only a few servers who does not have admins, and on those servers people make their own reports.

This got me to another point which I forgot to mention.

As you grab information straight from the HLstats 20 times a day with your bot on all players who has played on the server, it makes the HLstats lag. Before the bot was made, I had no problems with HLstats, which I do now. I belive Tolfx got a heads up by Kevin to remove a bot he made because it made the players lag more.
 
  • Informative
Reactions: Duck Dodgers

tres

Well-Known Member
Joined
Feb 10, 2020
Messages
42
Let some of the players who actually play at the time handle the report then. It's lazy.
So we should excuse players that are clearly breaking the rules just because the user who reported wasn't in the server?

As you grab information straight from the HLstats 20 times a day with your bot on all players who has played on the server, it makes the HLstats lag. Before the bot was made, I had no problems with HLstats, which I do now. I belive Tolfx got a heads up by Kevin to remove a bot he made because it made the players lag more.
1599078796136.png

According to Google Chrome's network, only 3.3 kB is sent to my computer, javascript runs client-sided so it should not impact server-side.
Since my bot uses the Python urllib libary which returns raw text, it should not download and render images or stylesheets.
It should not take that much bandwidth.
HLstats should not take a lot of computational resources, as it only stores user data and messages, with the ability to filter messages.
They do have a github https://github.com/NomisCZ/hlstatsx-community-edition

With the Tolfx one, I would have to ask him about that.
 
Last edited:

PNN

Trial Gameserver Admin
Staff member
Trial Gameserver Admin
Donator
Joined
Jan 27, 2016
Messages
877
So we should excuse players that are clearly breaking the rules just because the user who reported wasn't in the server?
Admins who has checked Hlstats and then placed a gag on SourceBans, which means they never was on the server, has been warned before. This is the samme issue, just with a bot.
 

tres

Well-Known Member
Joined
Feb 10, 2020
Messages
42
Admins who has checked Hlstats and then placed a gag on SourceBans, which means they never was on the server, has been warned before. This is the samme issue, just with a bot.
A player can report even if they aren't on the server.
What I believe is that admins can't gag or mute a player without being on the server IF there is no public report done.
The bot does not have elevated permissions, so this shouldn't really apply.
The bot processes the chatlog and makes a report. This report then later goes onto Player Reports where it is public and action can be taken by admins.
 
  • Indifferent
Reactions: PNN

PNN

Trial Gameserver Admin
Staff member
Trial Gameserver Admin
Donator
Joined
Jan 27, 2016
Messages
877
Right, I guss I have to wait for an admin to reply on this.
 

Addster

Well-Known Member
Donator
Joined
Jul 17, 2018
Messages
58
You can call people lazy for not doing manual reports. You can say your coded creation is less flawed than what people say. The thing is, it gets us nowhere.

I don't think the issue is about if the bot works or not. I think tres has shown it works in the ways it should. The issue should be if this is legal or not.

Just like PNN said, if admins are held up to standards of not digging into people's chatlogs, why not somebody's coded creation?
You could also relate this to chat filters. Chat filters are basically a limit on speech and can easily be evaded. How is this no different from that?
There are many other things I could ask, but I won't bother.

This begs the question: in what way is this allowed?
I'll leave it at that. Feel free to criticize.
 
  • Agree
Reactions: PNN

tres

Well-Known Member
Joined
Feb 10, 2020
Messages
42
Just like PNN said, if admins are held up to standards of not digging into people's chat logs, why not somebody's coded creation?
The bot generates a report that is posted at Player Reports, where it is public.
As described in [2][a], racism is strictly forbidden, and this goes with discrimination.
A bot generating a report and a player making a report should be no different, both aim to gag players who are obviously rule breaking.
As said before, the bot does not have elevated permissions to be able to gag users without posting a report.

You could also relate this to chat filters. Chat filters are basically a limit on speech and can easily be evaded. How is this no different from that?
This bot isn't perfect, of course, it will scan for obvious filter words. Even though some players can evade, the bot aims to at least be a good way to catch rule breakers.
You can take games with chat filtering, like the game ROBLOX, which has players evading chat filters.
Trying to put an anti-evade system or of the similar will increase searching.

This begs the question: in what way is this allowed?
This can pretty much be left up to what people believe.
I believe that a bot making the report for me to post should be no different as a player collecting data to report.
If I wiped the label of an auto reporting bot off, my reports would look just as legitimate as a player reporting.
 

Tolfx

Trial Gameserver Admin
Staff member
Trial Gameserver Admin
Joined
Mar 10, 2018
Messages
512
As you grab information straight from the HLstats 20 times a day with your bot on all players who has played on the server, it makes the HLstats lag. Before the bot was made, I had no problems with HLstats, which I do now. I belive Tolfx got a heads up by Kevin to remove a bot he made because it made the players lag more.
HlstatsX is already garbage to begin with, and there can be a huge delay by time and time, but I can agree there has been some delay these pass days. My bot tho wasn't grabbing from hlstatsX. It was grabbing from sourceban, to check recent comms and bans from users. Which I did poorly with puppeteer. As a result I checked each second to look for new bans/comms which gave a huge spike and Kevin told me about it, so I can make some improvements. Which I did and now it checks each 10 min (will probably do 30) with a simple http request, getting the raw html data. Tho I will probably stop checking since I gain no value from it, was only a learning experience.

I did however help Tres with this grabbing the chatlogs from hlstatsx, and used it myself for a while for testing purposing, tho I have no reason to check the chatlogs so is just a scratched code.

Now let me ask myself.. is this bot good? And I would say no sadly, even if I do support it. I think is unnecessary to have a bot like this, when it could be done by making a plugin for sourcemod to autogag those kinds of insults and etc. The bot lowkeys "spams" reports and an admin has to take care of it which seems also unnecessary, since the lack of activity of full admins (semi do your job). And I also don't like this sort of administration either, since I think every player should get a warning before punishment. But this is just my opinion about it, maybe the jailbreak team thinks something else who knows, good luck either way.
 
Last edited:

tres

Well-Known Member
Joined
Feb 10, 2020
Messages
42
To clear up some confusion (hopefully)
My bot does not run all the time. It doesn’t scrape HLstats all the time and doesn’t run for hours.
The bot runs only when I execute, and ends after it gathers all chat logs and evidence.
Estimated run time would be around 10-20 minutes, after that is all finished, the bot will stop scraping HLstats and Sourcebans.
The latest 2 times I have ran my bot was August 2nd and September 31st or 30th (Estimated)
It might be HLstats end, since HLstats sometimes lag or timeout even when my bot hasn’t been running.

@Tolfx I am thinking about making a single report thread for auto report bot to mitigate report spam if needed.
 

Roar

Well-Known Member
Donator
Joined
Nov 30, 2017
Messages
305
Bumping this thread for the ongoing issue, wonder if any admins can give some input into their thoughts on the auto-report bot? There seems to be some split opinions so far on this. If anyone missed it, the last run managed to report new trial admin @Gold simply for explaining in chat why he muted somebody - my question for the developer is, how many more mistakes and flaws in this system is it going to take until it either works efficiently and effectively, or is ultimately discontinued from operation?

Personally, the one major flaw I see, when faced with an automatic reporting system, is simply people catching on. I understand that currently the list of flagged words is rather small, with only 3 variations of racial slurs being included and scanned for, will this list be constantly updated for when people start using "1" instead of "i", "3" instead of "e", adding extra letters to key words to bypass the bot etc.? Is there anything proactive being done to prevent this from happening, or will it always be subject to what a non-staff member wants to include?
 

Madact

Gameserver Admin
Staff member
Gameserver Admin
Contributor
Joined
Nov 8, 2014
Messages
3,206
We also want to add that from now on auto reports are no longer accepted.

Any issues with this I advise you to speak to Kevin rather than starting this whole thing all over again. You'll need his specific permission. If you get that permission, then feel free to post these automatic reports again stating you got the go ahead from him.

With that, I hope that answers this question.
 
Status
Not open for further replies.

Users Who Are Viewing This Thread (Users: 0, Guests: 1)