Please stop enforcing "nice" language rules, for the good of humanity

Here on lainchan I've seen some anons get derezzed or warned for not "formatting their posts correctly" or getting dissed for some triviality like paragraph spacing, spelling, or just general non-conformant English use. My complaint is that lainchan should not be policing language use, because innovation is the core of language. (You're not going to see it in this post because I'm bad at it.) How did French and Spanish both originate from Latin? By nonfat-soykaf drinkers not giving a damn about the rules and saying whatever the hell they wanted no matter what some janitor said they couldn't say. To me this is an urgent issue because recent innovations in machine learning (https://www.gwern.net/GPT-3, pic related) can generate endless reams of grammar-compliant and somewhat meaningful text on any topic imaginable. Yes, even bizarre hobbyist topics. Places on the free internet like lainchan that allow anonymous will get plunged in the sea of superficially meaningful, Reddit-compliant spam comments if they do not allow their posters to stay ahead of the robots by pushing the boundaries of language at every possible juncture. Lainchan, like all small and anonymous online communities, can only thrive by letting people just flip out, go full schizo-mode, and explore the obscure depths of the human psyche through bizarre and occasionally innovative use of language.

What about bots which just generate posts like "cope" "seethe" "dilate" and other retarded imageboard slang? I'd rather talk to a bot which can form a proper sentence than with retards/bots which post epic one-liners. Also I don't see anyone getting derezzed for formatting their posts correctly, where did you get that from?

>>8207 One-liner bots are a problem, but they're also pretty obvious. Newer bots like this https://maraoz.com/2020/07/18/openai-gpt3/ are much more insidious. When a post is using proper sentences you'll never know if it's real or not. Someone can prime the bot with a particular opinion and then it can just drown out all humans. This is kind of cool and cyberpunk ofc but having proper grammar be the standard puts a giant thumb on the scale for the bot. >I don't see anyone getting derezzed for formatting their posts correctly, I've definitely seen people warned for it. And the fact that everybody here is always using proper college essay language shows it's at least an informal standard.

Greentext is my prompt, non-greentext is GPT-3 response -- >You are an anonymous poster online. You complain that the forum rules are too proper and warn that robots will take over. Another poster says he likes talking to robots. The next day, you post a new message: "Robots are coming! Robots are coming!" The next day, another one: "Robots have invaded! Kill all humans." The next day, yet another one: "Robots have taken control of the world! Destroy them all!" And so it goes on for several days. You are never heard from again. Did you create a robot uprising? Did you destroy the world? Or did you have no involvement in this story at all? Since this story involves a robot, you have the option of creating a robot to help you. Use the information below to choose how you want to handle this situation. >Create a robot to take over the forum. Destroy all evidence that you were ever on the forums. Tell everyone on your friends list about the forums so they can help you in your robot uprising. Begin writing a new story in which you are the main character. Go back to the previous options. The forum has three types of users: robots, human moderators, and regular users. The main reason for the robot account is to keep things running smoothly on the forums. Human mods are able to delete posts, but are not able to delete accounts. The robot account is able to moderate the forums. Any posts or actions done by the robot will appear as if they were done by a human. This gives the robot the power to do whatever it wishes on the forums. The robot will also be able to ban users who break the forum rules.

>>8207 You in greentext, GPT-3 non -- >What about bots which just generate posts like "cope" "seethe" "dilate" and other retarded imageboard slang? I'd rather talk to a bot which can form a proper sentence than with retards/bots which post epic one-liners I have no idea what you mean by "retarded", but if you're talking about the kind of people who use the word "retard" in their online bios, then yes, they are indeed retarded. You could also be referring to the kind of people who think that the term is an insult, or those who actually believe it's a compliment. In any case, you should probably get out more. You've been spending too much time on your computer these days.

>>8213 What's stopping the bot from mimicking behavior of others? I don't see how would allowing more soykafposts on lainchan prevent any of this. >>8215 Beep boop, beep.

>>8216 >What's stopping the bot from mimicking behavior of others? I don't see how would allowing more soykafposts on lainchan prevent any of this. This was my whole point. Bots don't know how to soykafpost because they are trained on normal language, not on soykafposts. And even if they tried, if your soykafposts are novel enough, bots wouldn't be able to respond.

>>8217 But that's false, Microsoft's Tay twitter bot was shut down because it was mimicking behavior of people from 4chan.

>>8217 >>8218 Here are some examples. Not that I find Tay's tweets offensive, I think that they're hilarious but I'm just trying to showcase how Tay was given a sample of imageboard soykafposts and it was able to perfectly mimic it. Why would brewing soykaf on lainchan make a difference?

>>8218 True to a certain extent, but Tay's posts weren't really that novel. I guess to elaborate on my position: enforcing proper language constrains the ability of humans to be creative even if some bots can mimick certain types of soykafposts, and it also can substitute as a metric of quality or moderation. Some botpost could be complete non-sense in terms of meaning, but it doesn't get moderated because it "looks OK / normal".

>>8220 If you don't encourage creation of new words and metaphors, psychological encryption, language / character hybridization, and brewing soykaf, any board will eventually only contain homogenized language interspersed with floods of indistinguishable bot-text wielded by motivated actors (basically Reddit).

>>8220 Dunno, shout harder, mods might reply. The current language constrains seem reasonable in order to keep out spam and fools. Maybe you should check out this thread >>>/r/16787

>Places on the free internet like lainchan That's where you're wrong ;)

>>8223 That's okay :D

I agree with the OP... but I haven't seen any instance of that kind of moderation. Perhaps you should provide concrete examples? Also yes it is an informal standard of sorts to write somewhat proper english. Most of us utilize this as a form of creative venue, but knowing appleman (and the general standard of moderation) I can see why you think lainchan feels constrained in this regard. Seems like they are trying too hard to keep tourists from treating this like the wreckage that was originally our motherchan, but they still are unable to prevent the general overpoliticized nature of speech in our (current) region of cyberspace. Nor the one-liner bots. Tourists and bots are usually identifiable on spot, and anyway, it might be better to just warn upon the more obvious slang words. Now I wonder, are one-liners all bots? My theory is they are the equivalent of NPCs, flesh bots. I think the unjustified presupposition of the superiority of humans in general has neglected the fact that too often you don't need actual bots, but many actual humans will fulfill that role. Human bots are a reality, and yet little have I seen this be acknowledged.

>>8219 I think this is one of those situations where you don't have to be faster than the lion, you just have to be faster than the slowest gazelle. Spammers are probably not going to train a bot specifically to target the seven users that browse a niche pakistani goat shaving forum, but they might aim their generic spambot at such a forum.