• 2 Posts
  • 1.72K Comments
Joined 3 years ago
cake
Cake day: June 11th, 2023

help-circle

  • I skim them all. The options I wrote are soft parodies of what each said. There’s a ton of problems with this model of governance, but I’m no voting member, so my opinion has no impact.

    That said, it makes it so a single posture has to be voted on in its entirety, which is dumb (as some legislative bodies of the world have proven). As a result, nuance is erased alongside with any sense of ideological vision or compromise, and instead it promotes vagueness in every action and glacial institutions that break slowly break apart.

    The very person who wrote the proposal that won reacted to winning with:

    I regret the loss of contributors (which would have happened with either option winnign [sic]). It is just too early to take a final decision on the mater [sic]. Let’s revisit this in two years or so, and let this mess of a GR be a warning for our future selves.

    The proposal was designed to win time without having to take any real stance while taking the smallest loss of contributors in the process. And thus it was written as the most milketoast nothing burger. When an infinite set of possibilities exists that actually do address real concerns like climate impact, social impacts, copyright and legal problems related to generated code, etc. in relation to the principles and overall values of the project. Instead, they are deferring the practicalities to pull request’s bickering.

    But again, I’m not a voting member or contributor. So my opinion is null. I still think the way the process is designed and conducted is stupid and prone to cause the eventual failure of debian via governance collapse.



  • The Debian vote was so stupid:

    Option A: Psychotic breakdown
    Option B: Ethical impracticality
    Option C: Political extremism
    Option D: Two contradictory things at once
    Option E: Something that doesn't prevent the other options but packaged with something nasty and unpalatable
    Option F: Sensible but ultimately vague and action-less common place neutrality.
    Option G: Nuclear option
    Option H: Do nothing
    

    It was such a brain dead exercise in futility. It was designed for the winner from the start.

    Nussbaun’s comment on the result is hilariously dull:

    On a very personal note, I regret that the winning option does not acknowledge all the legitimate concerns around AI: environmental impact, copyright and licensing uncertainty, the health of Free Software communities, and the strain of aggressive scraping on the open web.

    The GR results give us raw numbers about our opinion, and a direction, but we have a lot of work ahead of us to act as a community and understand how we can continue to accommodate the very large minority that would have preferred to ban or discourage AI.

    Lucas

    ¹ obligatory disclosure: the script used is AI-generated

    Yeah, the option that doesn’t say anything won. Who knew?…wait yes, mathematicians, they knew. The most important part of a survey is choosing the options carefully.



  • When you see that still the most used screen resolution is 1080 and even the very few people who are upgrading, do so only up to 1440. It is easy to see that graphics stopped being impressive a while ago. It seems that a long time ago people were still arguing FPS vs Resolution. FPS wins every single time. 24fps is fine for movies, it might add to the cinematic look. But in an interactive medium, FPS is king. Give me low latency and fast response times over anything else. People don’t notice when games look like shit, but they do notice when the gameplay is laggy or buggy. Most friendslop (compliment) games are very simple and basic in their looks, but still capture the gaming and streaming culture for months at a time. While AAAA games are now struggling to stay relevant beyond the launch week. It is the same trend as with movies. Blockbusters break records and then nobody ever remembers they even watched it. I guess that the algorithmic fragmentation of culture is finally catching up with us all.


  • Almost there but not quite. I’m not a lawyer, but have friends who are. The two examples are actually equivalent. Because, by principle, laws evaluate actions. Not consequences or intentions, those can be qualifiers in criminal cases, or discussed in civil cases. But in general the law prefers to deal with things that happened, not risk or what could happen. As it is far less ambiguous to argue that someone did something, than arguing whether or not something could’ve happened.

    That’s why pre-crime was such a big deal in the sci-fi movie Minority Report. Similar to thought crime, it is a tough ethical and legal discussion without a clear possible answer that doesn’t lead to violent authoritarianism.

    “I was speeding but nothing bad happened” is equivalent “I was shooting in his general direction not at him, I missed every time”. That’s still attempted murder and the former is still speeding. The crime is the action, not the result.


  • Bad news buddy. Ethics are politics. If anything politics is the warfare for the implementation of ethics. You can’t be ethical yet unwilling to enforce your ethics, and the only way to do that is to engage in politics. That’s just the only way to influence the behavior of other human beings.

    The worse part is that tech culture has failed to understand that all and any technology is part of the conflict to define politics. Whether you like it or not. Anything that involves a person besides yourself is political, because humans are political animals.









  • One thing to keep in mind is that, when it comes to LLMs, the models have not significantly changed in architecture.

    There’s been new experiments and advancements in architecture on neural networks, and machine learning for specific applications. But LLM, as they are being commercialized by AI corporations to the general public, have stayed relatively the same. Except for one thing. Increasing in size. Larger datasets, or more specialized datasets like with coding, and larger number of tokens in memory. This is why it takes such large data centers. It’s all been just brute forcing greater capabilities by enlarging the models.

    One of the things with LLM is that all the dataset influences the weighs and probabilities of the results. Even if the dataset includes a single event of a chain of words (think of the pizza with superglue incident), it can show up in the results eventually.


  • Your comment is so fit for purpose, displaying such a massive willful ignorance about the world in general. Because the biggest problem with asbestos wasn’t using it for home insulation. It killed millions of workers in manufacturing plants and mines. Its biggest industrial use case was for electrical insulation of high voltage transmission lines. Where it also killed hundreds of thousands of technicians in the us alone. Before that it was used for lamp wicks. Where it killed at least a couple of millions more over a century. Today, asbestos kills 250 thousand people a year, 12 to 15 thousand in the us every year.

    I supposed any tool can be considered useful if you’re ignorant enough about it or if you’re willing to lie without remorse.




  • dustyData@lemmy.worldtomemes@lemmy.worldCalling Collect
    link
    fedilink
    English
    arrow-up
    9
    ·
    21 days ago

    Even most landlines today just relay to a fully digitalized VoIP infrastructure. The new stuff is just way more economical to run, even if the residences still run kilometers of old copper wire, the hubs have changed to microwave antennas or satellite uplinks, as they are much cheaper to maintain.