God Emperor Doom Programming shitpost #3: A case study of Minds, and the next stepping stone

Programming shitpost #3: A case study of Minds, and the next stepping stone

(Originally posted on Minds.com)
Jul 12, 2021

What have we learned from this grand experiment? Money ruins everything. At least it does when you throw it at beggars and harlots while screaming "notice me senpai!" Some comms platforms are overrun by thots and their paypig simps, some have plain old patronage and grifters, some have ad revenue sharing and the clickbaiting and censorship it brings. Generally you have to do videos or livestreaming to prosper. Minds aspires to be a digital economy where all kinds of content creators can prosper - vloggers, bloggers, writers, artists, photographers, composers, memelords, and shitposters. It only takes a minute to sign up, and if you learn to code, you can do it every minute of the day and night. Steal other people's content, post it here, and some dumbass will toss a Minds token in your guitar case. Repeat several thousand times in parallel, funnel the tokens to your paid account, and cash out in Ethereum. You didn't ruin the system, you brought it to its logical conclusion.

How do you prevent this? First take money out of the system. Ok, but users can still make deals that promote the same moneygrubbing behaviors. Ban them, they'll use another account. Auto-ban them with AI, you'll catch a lot of legit users in the crossfire. Demand phone numbers or ID, you'll lose everyone who cares about privacy.

But remember, we want decentralization. Total user control, no admins or mods, no bans, no censorship. So it's up to each user to decide who to block (or segregate into different feeds/categories) and who to communicate with in the first place. There's an endless horde of spambots out there on the internet. You don't even answer your door when they knock. You don't even have a door. If they wanna reach out and touch you they need to be introduced to you IRL, through a mutual friend, or in forum or something. Maybe there's a pool of mutual acquaintances who can request to talk.

Decentralization is hard, so I would build a comfy little chat server first; that's easy. Then I'd add bulletin boards, mail, images, file sharing, etc. And if I get enough users to need blocking, filtering, etc, I'd start working on that - and just as importantly the UI for it. Maybe you don't wanna block Rambling Joe, you just wanna reduce his messages to a 4 pixel font, lol.

If I had to build it in a hurry because everything else is going to shit right now, I'd make it web-based.

Later, if I've got a good working prototype with lots of users on several independent servers, I'd tackle hard problems like decentralization, encryption, and cutting ties with the web.

*I would NEVER add follows, subs, likes, stickers, or buttons. Okay maybe in the beginning by popular demand because everyone expects it. And polls - they're the shit. But the uncomfortable truth is that likes/follows/subs are fake and gay. They turn everyone into simpering little attention whores. The good news is, you can't have them in a fully decentralized system. You can know your 'mutuals' count, and you can share it, but no one's gonna believe you and there's no central arbiter of trust to say "yes, it's true, this douche has 458k subs."

Part 2: Gatekeeping

Gatekeeping, how will it work? Assume we have a decentralized platform based on encrypted blockchains, like Ethereum and Web3, but probably just for information storage, not transactions. The designers could learn from cryptocurrencies but I'd keep money separate at least until the dust settles. You have member nodes (computers/devices) passing along copies of encrypted data by request, a system for searching public data stored on other nodes without copying it all, and on top of this, the usual file/video sharing, chats, forums, wikis, etc. When you open a link (a crypto hash instead of an http url) your browser sends out a peer-to-peer request, receives and decrypts the data, requests any additional data needed, until you have everything. And you get a full copy to share/archive if you want, like BitTorrent not YouTube. Groups/forums are basically blockchains... you get a link to the blockchain, retrieve messages to read, append a new link to post. I don't know how cryptocurrency consensus algorithms work; maybe forums need something completely different. Maybe the group owner has sole power to approve or reject posts. Anyone who disagrees can fork the group and disgruntled members can follow his blockchain instead. What about more permanent info - decentralized wikis, maps, etc? Instead of one "fair and balance" version, everyone with an opinion about a topic has their own version. Man, I would delete so much crap from wiki articles and just link to the best sources. If you want alternative viewpoints, search and compare. Maps are better suited to decentralization. Markers and paths are just map coordinates, so you can request them from many different sources, show them all on the map, and use algorithms to detect discrepancies (same thing in 2 different locations, two different things at same approximate location, etc) and warn you about the confusion. You can give preference to reliable sources/gatekeepers and ignore spammers and idiots.

Part 3: Advertising/Discovery

Could we have advertising in a voluntary, anonymous, encrypted, decentralized platform? One way is to simply post your ads in a forum where that sort of thing is welcome. If you annoy people they'll mute/block you. Someone could create an advertising service that people willingly sign up for to discover new things. Assume it's just a user (bot) that sends text/image/audio/video messages to you when you sign up, completely anonymously. You choose which categories of ads to receive... i.e. video games, scifi novels, and 5 flavors of porn. It sends you ads with like 3-digit codes in them, so if you actually watch/listen/read the ad, you can reply with the code to claim a small reward, as a thank you for giving feedback. You can use your rewards to run your own ads on the service, or redeem them for a discount when you buy from an advertiser... hopefully that can be done anonymously. Naturally, scammers will try to game the system... advertisers advertising in the wrong category, prohibited ads (ad services generally don't want to be Silk Road), viewers typing the codes as fast as they can, bot farmers running thousands of fake viewers with AI... so you get rewarded for reporting violators; you get diminishing rewards for going too fast; and advertisers get rewarded for not making it too easy or difficult to find the codes. The ad service can decide how to do all this, and if viewiers/advertisers don't like it, they don't have to use it. How would the ad services make money off an internal token system? Donations from happy advertisers and viewers, lol. They could also sell views for money, but if you see too many big corporate ads you could just unsubscribe. They wouldn't own the platform and 80% of the ad market. Will it work? Probably not. It would be difficult to prevent bot farmers from gaming the system, like they do here on Minds. Anyway, whatever you do, don't build shitty advertising into the platform itself. Scammers will game the system and no one else will use it.