• 0 Posts
  • 5 Comments
Joined 1 year ago
cake
Cake day: June 28th, 2023

help-circle
  • It’s possible that it eventually ends capitalism, or at the very least forces it to reform significantly.

    Consider that the most basic way a company can obtain profit is by extracting as much surplus value as they possibly can, i.e spending less and earning more. Extracting high surplus value from human workers is easy, because a salary doesn’t really depend on the intrinsic value of the service a worker is providing, but rather it’s tied to the price of that job position in the market. Theoretically, employers can all agree and offer lower salaries for the same jobs if the situation demands it. You can always “negotiate” a lower salary with a human worker, and they will accept because any amount of money is better than no money. Machines are different. They don’t need a salary, but they do carry a maintenance cost, and you cannot negotiate with that. If you don’t cover the maintenance costs, the machine will outright not do its job, and no amount of threats will change that. You can always optimize a machine, replace it with a better one, etc. but the rate at which machines get optimized is slower than the rate at which salaries can decrease or even become stagnant in the face of inflation. So it’s a lot harder to extract surplus value from machines than it is from human workers.

    Historically, machines helped cement a wealth gap. If there was a job that required some specialization and therefore had a somewhat solid salary, machines would split it into a “lesser” job that many more people can do (i.e just ensuring the machine is doing its job), driving down salaries and therefore their purchasing power, and a specialized job (i.e creating or maintaining the machine), which much less people can access, whose salaries have remained high.

    So far, machines haven’t really replaced human workforce, but they have helped cement an underclass with little purchasing power. This time, the whole schtick with AI is that it will be able to, supposedly, eventually replace specialist jobs. If AI does deliver on that promise, we’ll get stuck with a wealth distribution where a majority of the working class has little purchasing power to do anything. Since working class is also the majority of the population, companies won’t really be able to sell anything because no one will be able to buy anything. You cannot sustain an economic model that impoverishes the same demography it leeches off of.

    But there is a catch: All companies have an incentive to pursue that perfect AI which can replace specialist jobs. Having those would give them a huge advantage for them in the market. AI doesn’t demand good working conditions, they don’t undermine other employees’ loyalty by unionizing, they are generally cheaper and more reliable than human workers, etc. which sounds all fine and dandy until you realize that it’s also those human workers the ones buying your products and services. AI has, by definition, a null purchasing power. So, companies individually have an incentive to pursue that perfect AI, but when all companies have access to it… no company will be sustainable anymore.

    Of course, it’s all contingent on AI ever getting that far, which at the moment I’m not sure it’s even possible, but tech nerds sure love to promise it is. Personally, I’m hopeful that we will eventually organize society in a way where machines are doing the dirty work while I get to lead a meaningful life and engage in jobs I’m actively interested in, rather than just to get by. This is one of the possible paths to that society. Unfortunately, it also means that, for the working class, it will get worse before it gets better.


  • I didn’t know about Nostr, and honestly, thank you for letting me know. I find the idea behind the protocol to be pretty interesting, but wouldn’t use it myself. I’m not keen on the idea of removing moderation altogether.

    However, I do think that it’s about time we rethink how moderation works on the Internet, and the fediverse should be a good place to start doing so. Perhaps my biggest gripe with the fediverse is that moderation works exactly the same as in corporate social media. Right now, moderators are picked under discretion of whatever the criteria of the admin are, and they are not subject to “the will of the people” so to speak. If a mod or admin acts in bad faith, the only recourse for the rest of the users is to leave, and maybe setup your own instance if you have the technical know-how. And while corporate media admins are somewhat constrained by investors, fediverse admins don’t have to respond to anyone. Which is better than being bound by investors, but here, admins can and do take harsh decisions on a whim without having to justify anything to anyone. Which is honestly not a good thing.

    So, while I imagined the fediverse as some network of interconnected small, self-managed communes, what we actually have is a network of petty fiefdoms, some of which do listen to their users even though they are under no obligation to do so, and others outright don’t. I don’t mean to say that centralized services are better at this, but in the end I’m having some of the same problems regarding arbitrariness of moderation and admin decisions here that I had on Reddit and Twitter.

    I see the fediverse as the future of social media, but not in its current form. The way it currently works keeps us bound to drama and petty feuds between admins of instances, and that is unavoidable while large fedi platforms are hosted by single people or very small groups of people. Perhaps the way that this could be avoided would be by using a protocol that enforces decentralization of hosting, like Nostr does. I imagine it would work sort of like a torrent, where we are all sharing and hosting the instance or the communities we use, whether completely or only partially. Or perhaps an instance is made out of multiple relays which are hosted separately. This way, we wouldn’t have issues such as admins unilaterally defederating instances because of a disagreement or stuff like that, since we’d all be admins in a way.

    I wouldn’t want to do away with moderation, but decisions such as who gets to be moderator, who gets to keep being moderator, and who we ban, fed with or defed from, is consulted via democratic process enforced by design. Otherwise, it’s not going to be meaningfully different from centralized media once the big instances become big enough.