• 43 Posts
  • 414 Comments
Joined 2 years ago
cake
Cake day: June 20th, 2023

help-circle

  • I see a lot of people answering what the Republican leadership seems to believe and do. This is very different from what the average Republican voter believes.

    Let me cover just a couple issues that drive an otherwise functional person to vote Republican:

    1. abortion. There are a shocking (and tragic) number of otherwise reasonable people who get a strong ick response to the idea of abortion. They rarely research the issue. This is an opinion so immediately visceral that they believe it is a moral law (and there are many communities that strongly reinforce this belief). I know of many folks who see the Republican corruption and the damage it’s doing, but can’t get through their anti-choice gut feeling to vote blue.

    2. economic interest. There are a nontrivial number of people who could lose a lot of money depending on how democratic policies are implemented. The left is a fractured mess, so no particular implementation is guaranteed. But if you keep the system around, historically you won’t get harmed specifically (we all get leeched to death slowly instead). Think rent control for a family whose retirement depends on 3 rental properties. The grandkids will vote Republican to, they believe, keep Grandma solvent. (This wouldn’t be an issue if large sweeping reforms were on the table. They aren’t. Also these calculations are often vibes based, because who has this kind of data.) See all the incentives around NIMBY Democrats. Some previously union areas fall under this; globalization policies felt like they destroyed their communities.

    There are more. And polling will tell you about them. No need to ask other leftists (we call this an echo chamber).





  • They don’t say that the random answer is chosen uniformly (though that is the norm in the field). If we relax that, then we’re putting a distribution on these where we want:

    P(correct with distribution (a,b,c,d)) = some value shown on A,B,C,D

    I don’t see any assumption that we will pick using that distribution, so I think this avoids the recursion.

    Unfortunately this has too many solutions. If you put a total of 0.25 weight on A and D, then the rest does not matter. If you put 0.5 weight on C, again the rest is irrelevant.



  • I apologize for not reading you carefully enough. Indeed, I simply disagree with your application of the labor theory of value, to the point that I think your claims in this specific case do not make sense. (Some of the arguments made here I think are strong reductions-to-absurdity against the labor theory, but I will try to explain my objections from within the theory.)

    The value I’m referring to is the value inherent in a production of a commodity that originates from the raw materials and labor that workers put into it. I’m talking labor value. It’s the value of the grain that originates in the workers toil and the raw stuff.

    I think you can salvage many cases of this theory of labor+materials by picking good boundaries for what a ‘product’ is. What the market chooses to say it is selling to you is generally lies, which we shouldn’t trust. What it actually is selling to you is often a large amalgam of things. We should consider the value of a whole bag of goods that the consumers/world can interact with because of the purchase. When you buy a train ticket, we do not judge its value by the quality of the printing (or, more generously, the labor+materials of the chair you sit on). It’s value comes from all of the train engine and the train car and the tracks they ride on and the stations they visit and the people who run them. In this lens, I think we should not consider a game as only the programming+assets made by the dev team. A game is thus, at least:

    • the infrastructure to sell it (storefront, payment processing, reviews, recommendation algorithms, marketting, hosting of data, etc.)
    • the hardware to run it (consoles or PC, drivers, operating systems and chip specs, special controllers, etc.)
    • the programming and assets (most of what you are talking about when you say game, but also cloud saving and account management. How a game manages saves when the power goes out is very much part of the game.)
    • the infrastructure to run it (multiplayer servers and connection protocols, anticheat and moderation systems, friends systems, in-game monetization systems, and the internet infrastructure probably counts too.)

    You could maybe argue Valve creates value in the production and maintenance of the commodity that is Steam’s infrastructure and sell it at a fair price. But in this context, the whole point of that infrastructure is to realize the value created from the labor of developers, making it extractive in nature.

    I cannot accept a theory of value that says tabletop simulator’s value is entirely independent of if you can play multiplayer. It is, to me, a multiplayer game. I am reasonably certain that the game devs don’t host that server, did not write the server code, and do not manage connecting different players together. That’s all part of the product, and I think it’s a part that valve should get credit/payment for.

    I note that when game devs did not generally use steam, valve made their own (excellent) games. I suspect if people started leaving the platform, they would return to it. I think this refutes the claim in this quote. The infrastructure is so good that it genuinely improves many many games. (Now this has extended to marketting, so that even games that work great from itch.io also want to be on steam. I agree that these network effects are much closer to rent seeking and don’t add value under the labor theory. But I have not seen a strong argument that valve has been anti-competitive with this privilege, and I think it is just wrong to assert that valve does not contribute to many of the games cited in this thread.)

    If you package a game as not just the assets and code, but also the things that make it possible to use those in the way the game dev intended, then steam is selling a large portion of each game. Every time I run a steam game, I am running a substantial number of lines of code written because valve exists, often more than the number of lines of code written by the game dev. If I bought the game elsewhere, this would often not be true.

    Steamdeck does not increase the economic value of any other product (though, I love my Steamdeck)

    Infinite effort and using up all the gold on earth, spent making a game that can only be run on temple OS installed into a vintage N64 in orbit around mars, is entirely wasted. The game would have no value because there are no consumers: it cannot enter the market. The labor theory of value only makes sense for things that are, in fact, sold. To take another example, train engines are valuable because train tracks exist and vice-versa. There is no service to sell with just one of the two. We live in a society.

    (I do not intend to argue valve is perfect; in ways and cases they are a monopoly and have behaved as rent seekers. But I think your claims are too strong/sweeping.)


  • I think I understand your position and disagree on many games (as others may have communicated better). I think most games I play are a better deal because steam and valve exist. Cloud saves and multiplayer setup both have definitely quadrupled the value of many many games for me, it is not close.

    Also, valve created a new hardware to increase gamer access via the steam deck. Those sales and that user base both add value for devs. Similar argument for Linux, except valve doesn’t even rent seek on most of that. The drivers are for everyone.



  • My gut says capitalism. So I would be extremely interested in examples where:

    • There was a large open community, hosted not-for-profit
    • The community is now thoroughly dead, with no serious capitalist alternative or clear successor

    4-chan isn’t dead, is still pretty open, and doesn’t have a real competitor for what it is. Certain forums and chatrooms would qualify if reddit didn’t exist (but reddit really clearly killed several of these).


  • I think the first observation is basically true, and the generalization: “some people are not working with your world model or goals. Electoral strategy and persuading arguments should be evaluated on their merit, not the person giving them.” I think is also true.

    In that setting, Mamdani is a useful test case (‘is this progress? No? Then your model is alien to mine!’). I would prefer counter arguments (over examples) to purity tests, or a readership aligned enough to downvote them. I would especially like to not have to put my interlocutor in a bucket (‘are they tankie?’), because that seems inefficient and often alienating. Folks often parrot points they don’t understand nor deeply hold.

    I think the second observation is also basically right. Also remember the folks who agree with doomer sentiment, because it matches the feeling that all is going bad, but don’t know why. It may be worth writing and arguing to persuade them (even if the person who argues back is a lost cause).

    Would be interested in data on how many folks hold beliefs like “Mamdani is still far right”, how many of them vote, and how many of them can explain their position. My guess is that this is a fair few people, but only a sliver of them act on it or understand if deeply, but data would be great.