• 2 Posts
  • 277 Comments
Joined 1 year ago
cake
Cake day: June 24th, 2023

help-circle
  • When writing my previous post I had started writing a list of suggested strategies; but I changed my mind about posting that. I’m not a member of Mozilla. I don’t know what particular challenges they face, and my expertise are not in not-for-profit fundraising. So although I do have ideas, I don’t really want to get into a trap of trying to defend my half-arse ideas against people picking them apart. It’s beside the point. The point is just that it is achievable, as evidenced by other organisations achieving it.

    I will say though that they could at least just mention on the Firefox ‘successful update’ page that Firefox is supported by donations, and give a link. A lot of people really like Firefox; and I think that if Firefox asked for donations, they would get more donations.












  • I don’t think its viable to post on Facebook (or whatever). with the goal of getting people to stop using Facebook. People on the site will quickly disregard whatever arguments put forward, because it is very hard to maintain credibility while you are on the site that you are arguing against. And if you do make an effort to maintain your credibility there, then you can end up having the reverse effect - in the sense that you are now a valuable contributor to the platform, and people will like and respect your contributions while staying on the platform.

    Criticism one platform is more likely to be taken on-board if posted on a different platform though. For example, a lot of people on Reddit argue against Facebook - and I reckon it probably has an effect the people who read it. But obviously it isn’t the ideal target audience. You really want to arguing against Facebook to people who are actually on Facebook!

    I can’t really think of a good way to get people to stop using particular social media en mass. (I’d written a couple of paragraphs about general strategies for changing people’s minds; but I don’t think it was helpful enough to bother reading.) I suppose the best option would be government regulation to undermine the targeted advertisement revenue streams. If these platforms were forbidden from collecting and using personal data for advertising, then the incentive for content-churn would be reduced. Without the money, the corruption would no long be self-sustaining. So political action to support strong privacy laws is probably the best way. Aside from that, probably the best way is to degrade the quality of the platforms. Don’t contribute any content to them. Don’t give them any kind of credibility by linking to them (for any reason, even negative attention). If you feel like being chaotic, I suppose you could create spam bots to just make the place worse, but that’s probably not worth the effort. Actual spam-bots are common enough anyway.


  • The normal Windows task manager’s ‘end task’ button just politely asks the app to close - but then later will tell the user if the app is unresponsive, and offer to brutally murder it instead.

    There is also the sysinternals Process Monitor, which is basically ‘expert’ version of the task manager. Procmon does allow you to just kill a task outright.


  • I largely agree, but I will say that it isn’t only about a financial safety net. AI corporations are using huge trawling nets to pull in the work of everyone in the world, and then resell it in a convenient box. The fact that the profits will be unevenly distributed is only one negative side effect. Because just like ocean trawling, the other side effect is that it will leave the ecosystem damaged and diminished.

    Note that the comic in this case is Penny Arcade. Those guys are part of the first original wave of web-comics. They are pioneers and veterans. Their regular blog posts are a level-headed contemporary commentary of the state of the internet and of games. The website is amusing, but it is also a good historical document. And although their huge success is largely due to luck of their timing, and perseverance; they have used their success to make great contributions well beyond just the comics. (I’m thinking mostly of their charity “Child’s play”, and the various PAX gaming expos.) So that’s the kind of value we risk losing, even if AI profits are shared ‘fairly’.

    In the comic, (and in a couple of recent blog posts), they are basically concerned that their work is being used without their permission to train AI to mimic their work, and the work of other artists. Partially this is about money, but it is also about clarity of communication. The comics, and their blog have always been a way of communicating their thoughts and chronicling history. And a flood of low-effort AI replicas can dilute this to a level of pointlessness.

    And its a similar situation with all artists, with some artists being far more vulnerable than others. Artists generally are not simply drawing stuff to get paid. They are trying to communicate something about the world. So this isn’t only about getting paid for art. It’s about being able to contribute meaning. With AI being produced at a rate far far higher than human art, the signal-to-noise ratio will drop sharply.







  • AI is a lot like plastic:

    It is versatile and easy to use. There are some cases for which it is the highest quality product for the job; but for most cases it is just a far cheaper alternative, with bit of a quality reduction.

    So what we end up with is plastic being used a lot, to reduce costs and maximise profits; but mostly the products it is used for are worse than they would otherwise be. They look worse. They degrade faster. They produce mountains of waste that end up contaminating every food source of every animal in the world. As a species, we want to use it less; but individual companies and people continue to use it for everything because it is cheap and convenient.

    I think AI will be the same. It is relatively cheap and convenient. It can be used for a very wide range of things, and does a pretty good job. But in most cases it is not quite as good as what we were doing before. In any case, AI output will dominate everything we consume because of how cheap and easy it is. News, reviews, social media comments, web searches, all sorts of products… a huge proportion will be AI created - and although we’ll wish they weren’t (because of the unreliable quality), it will be almost impossible to avoid; because its easier to produce 1000 articles with AI than a single one by a human. So people will churn junk and hope to get lucky rather than putting in work to insure high quality.

    For individual people creating stuff, the AI makes it easier and faster and cheaper; and can create good results. But for the world as a whole, we’ll end up choking on a mountain of rubbish, as we now have to wade through vastly more low-quality works to find what we’re looking for. It will contaminate everything we consume, and we won’t be able to get rid of it.