Yes! This is a great idea. On the topic of switching back to the actual topic, do you have any comments on how you feel about using automation to flag code? I see a lot of people disliking it (me included) but It is always a good idea to see everyone view point.
Perhaps you have shared already (but there is 1000+ replies and it is hard to keep track without monitoring this thread 24/7)
Using automation to flag code is wrong. The automation for text is difficult, because text is a very human thing. Surprisingly, machine code is even more human, and requires more context, due to logic on how code functions and why a developer might have put a seemingly bad string out of context.
Edit: By machine code, I just mean code, not literal machine code that you would find the next level up from binary code
This appears to be related to their push for more security on the platform. Hence the new HTTP Plugin Permissions Update.
The push for more security could be related to the Meep City incident that occurred recently. (If you don’t know what I’m referencing just search up “meep city hacked 2020” or something)
Oh yes, I think it’s err… Not great. I say we need to do manual, because if it’s in automatic, it’s gonna be over 90x easier to get someone banned. Which I stated in my overall opinion on this subject. If it’s human, they can probably tell what’s going on, and if it’s fair to use the swear words. Like extra chat-filter and etc etc. But I also don’t really want humans either. As they get access to person information such as API Keys, and Bots on certain dizzy servers. So I really just think this shouldn’t be released overall.
You do not get immediately banned if you are flagged. Upon being flagged you are subject to manual investigation with a “specialized team”.
I am worried said specialized team will not be able to handle the sheer amount of stuff they would have to go through. Depending on the code it could be very hard to tell what it actually does. Especially if it becomes obfuscated.
I think automation for moderation is cool, but I don’t want things that I choose to keep private to have to get searched.
Wrong, read what NickoSCP said. This system is very broken. Like just think of this. You make an extra chat-filter to prevent free robux scams, you go grocery shopping (or anywhere), and come back to being terminated… Immediatly. Banned for no fair reason. So that is wrong, and even if there is a team, they are not very good at their job as Nicko still got banned for unfair reasons.
What about obfuscated code? Do we get punished for that?
Exactly, so don’t touch it!
The parts of my code that are visible to players of my game can be touched and reviewed, but do not touch the rest of my code.
I am writing my code on the way how I want to write it. If I want to put some notes in it for myself (which may contain numbers or other things that are filtered in the rest of Roblox), I should be able to keep them in.
As @wevetments said above:
Is this really needed? There are many better ways to implement this, for ex. someone reports the game, it gets scanned, then sent to moderation.
Alright, I just realized a major issue.
How would this be handled if a malicious plugin were to inject/plant backdoor code that contains profanity?
Synapse Xen a popular obfuscator contains lines of profanity in the code. Would the person NOT AWARE that this script exists in the game be liable/responsible for the script containing the language that was implemented by a malicious plugin?
I’m assuming that there is no way to check if a script has been implemented by a developer or plugin.
If that’s the case then new coming developers are probably the most likely targets of this moderation.
So it would be safe to assume that we have alot of false bans ahead of us because of the MASSIVE-MASSIVE malicious model/plugin issue that already exists.
And if I’m gonna speak about plugins one more time, implementing plugins as a marketplace to profit out of made the backdoor issue increase 10 fold due to the fact that people would get the purchased plugin, copy the source and implement its backdoor code and release it into the marketplace for free.
You essentially introduced a unstable and unfinished system into ROBLOX with no insurance/protection of the seller and no way to actually stop people from spreading the sources of the plugin. As far as I am aware there was no effort put in to actually assist developers wanting to profit from the tools they provide the community.
It’s not just useless, it’s an invasion of privacy.
Wow, The responses doubled since I check it last time. Roblox you really have to reconsider about this base on chaotic controversy.
While most of the replies are criticizing about how this should not be implemented, I have to be honest here, the initial point and the mindset of this does make sense for Roblox, as a kid friendly platform. It makes sense that Roblox should not allow any inappropriate content that is built on the platform regardless it’s visible to the public or not. I did see a Free Model with script which included links to inappropriate websites with adult content before, despite the fact that it was 2 years ago. However, the way that Roblox is running this is not appreciable.
Firstly, transparency. This has been ran for a month without notifying developers? This is not a good initiative to implement such controversial system. The output of this can be seen by these 1k replies in this thread.
Secondly, lack of information and misconception. I’ve heard a developer got terminated because a script which filters bad words. I’m not sure what happened to him/her now, but surely this will leave us all a bad impression. It’s also a major concern on the “professional team”. We know nothing about them, we can choose not to trust them because it’s our code and maybe they included some sensitive data as mentioned in threads above.
Last but not least, this is not a major problem. This might not be directing to the OP, but it makes sense right? We all know there’s a room of improvement can be made regarding on the Roblox moderation system. Not only this moderation harms player sometimes, it also impact developers in such ridiculous way. A developer can get banned for 3 days while uploading a harmless image. Allow me to discuss and talk about this here please. I uploaded 2 versions of a same audio because the first one was too quiet to hear, then the first audio got me banned AFTER A YEAR while the second one remains normal. This problem is more urgent that moderating something private and cause no urgent harm to anyone, right?
I don’t really need anymore constructive argument, most of the threads has already covered and represented me. I hope Roblox will take a better action in the future.
I did see a Free Model with script which included links to inappropriate websites with adult content before, despite the fact that it was 2 years ago.
If we want to get specific and very technical. The botted models that contain malicious scripts which are most likely obfuscated with Synapse Xen, contains vulgar language as it’s a default with the Synapse Xen obfuscator.
Jesus Christ, you can’t add this update, no you can’t !
Every developer’s code is his code and nobody has the right to see it, personally I don’t want any human from your team to check my code, and considering how bad your moderation system and bots are, obviously I could get my account banned for nothing. What if an exploiter gets every code leaked just because of a “security” issue? We all risk our games to get exploited and maybe stolen, considering you will also check the server-side code! I highly suggest improving your current moderation system and forgetting about this horrible idea.
This update has existed for a long time, I for one have always assumed Roblox looked at my code so it came as a shock to me that others didn’t think Roblox would look at our code.
Well, ROBLOX liked to enforce a phrase/policy of your code is your code. Assuming they are saying we own the rights to what we make. But I’m guessing the right to keep my content private from other people isn’t a thing and now I would have to worry about my API keys being compromised. I highly doubt you would need a “specially-trained” team to decide whether code has intentional profanity or not. But who knows you can get banned for pretty much anything now a days.
EDIT: Ya now that I think on it, the “team” is probably the regular moderation team that does pretty much everything. Calling them “specially-trained” is probably to put minds at ease. But I still have no trust for the team to keep code private and safe while reviewing it. I’m going to be looking into obfuscation options for the future due to this.
Owning your content is not the same as being the only one with the right to inspect that content. Roblox is allowed to inspect your content as per TOS (doesn’t mean they own the copyright to it, you still own it).
Yes, because they obviously want to lie to their community. Should I fetch you the tinfoil hat from the cupboard?
Regular moderators who review assets are not technical and wouldn’t be the ones reviewing the code.
Ditto. I am also planning to move away from Roblox as soon as possible. I intend to release the game I am working on currently, and once I do, (regardless of whether I make profits on Roblox or not) I will begin rewriting that same game on Unity (which I am already very familiar with, so yay). The hope is that I make a decent chunk of change on Roblox so I can fund or at least kickstart my ambitions elsewhere. I am sick of the negligence of Roblox in their handling of developer affairs.
Does this system automatically flag utf-8 encoded strings?
print("\x48\x65\x6C\x6C\x6F\x20\x6D\x79\x20\x6E\x61\x6D\x65\x20\x69\x73\x20\x57\x65\x62\x6D\x6F\x74\x69\x6F\x6E")
Hello my name is Webmotion
Alright, I should of probably added this is speculation which is a my bad. But you don’t need to actively become condescending to me because I have a different opinion. But if we were to move on to a different topic(most likely more important) instead of my speculation, this is probably my biggest concern then.