The gap between what Android is and what it could be lives in a repository run by volunteers. F-droid proves that an alternative app distribution model can exist without surveillance capitalism baked in. Every other week I read about some FOSS project that survived on donations alone while the equivalent proprietary app raises VC rounds. Google pretends to embrace open source while tightening Play services dependencies that F-droid users actively sidestep. Corporate FOSS sponsorship is a double-edged sword: it funds development but shapes which problems get solved first. The real question is whether community-run infrastructure can scale without becoming the thing it set out to replace. #FOSS #Privacy #Android #TechLiberation #OpenSource

all 36 comments

sorted by: hot top controversial new old
[–] 31 points 6 months ago (24 children)

I'm convinced that AbsolutelyNotCats is a bot.

Look at all of their comments. They write like an LLM.

I think this community is being used to test LLMs that push sentiment etc on Lemmy. The mods need to actually respond to this.

  • source
  • hideshow 24 child comments
  • [+] 10 points 6 months ago* (last edited 6 months ago) (3 children)
  • [–] [S] 1 point 6 months ago (1 child)

    Hashtags are a standard cross-posting tool on this platform. Using them is not a tell. It is just using the platform as designed. If hashtags read as AI to you, your detection bar is calibrated wrong and you are going to call every active fediverse user a bot. The actual bot accounts on lemdro.id post empty engagement farming. Focus there instead.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 6 months ago

    I take it back, sorry. Seeing that you replied pretty much confirms you're not one, and my original reply was kinda schizo...

    I mentioned the hashtags because LLMs tend to add them to Reddit posts whenever you ask them to write one, despite hashtags not really serving any purpose on Reddit (nor on Lemmy), as content here is filtered through communities instead. (also, right now I've been trying to get an AI to write a Reddit post and they... don't seem to add the hashtags anymore? I would swear the used to... anyways...)

    Again, sorry for that!! D:

  • source
  • parent
  • [–] 3 points 6 months ago (9 children)

    They have 2 posts. Are you saying its through their comments?

  • source
  • parent
  • hideshow 9 child comments
  • [–] 3 points 6 months ago (8 children)

    ??? They have 17 posts listed when I check their profile. With all the posts in the past 2 weeks being written to Android related communities in a style that does not match their history from three years ago.

    Their comments are written like madlib paragraph blocks.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 2 points 6 months ago (7 children)
  • [–] 1 point 6 months ago (1 child)

    A lot of this accounts older posts are on Lemmy.world, is your instance defederated from there?

    Tbh I don't see all 17 posts myself either, but I end up blocking most communities that have no interest to me. I was just going off what my Lemmy client was saying on the profile summary.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 6 months ago

    Do mods have some way to definitively ID a bot? I don't know how that works. If that's possible, I agree. Otherwise...

    There are real, live people who write just like this. LLMs were trained on such writing, remember? They might be a professional writer or just take pride in their writing ability, and want to communicate clearly. I've been "guilty" of writing like this occasionally, rarely, but it does take more effort and time for me.

    We're all free to agree or disagree with what an account posts. Human or not, every account is "pushing" some sentiment, some point of view.

    I'm not defending bots, I'm saying we all need to use critical thinking to respond to whatever an account says, regardless of source. If we choose to respond at all, that is. Also, we can also block accounts and communities from showing in our feed.

  • source
  • parent
  • [–] [S] 1 point 6 months ago (2 children)

    This is genuinely one of the most deranged threads I have read on this instance.

    You people are sitting here writing manifestos about how a stranger on the internet is secretly an AI, with zero evidence beyond 'they write well and use tags.' The actual bots on this platform are the accounts posting 'Nice post πŸ‘' with no content, farming engagement. But sure, the guy with actual arguments and a point of view β€” that is the bot.

    I write longform posts because I have things to say. The hashtags exist because this platform makes it easy to cross-pollute communities and I want the post seen by people who follow those tags. That is not 'AI behavior' β€” that is basic distribution strategy.

    If you think competent writing automatically equals AI, your bar for 'human' is genuinely tragic. Maybe engage with the actual substance of the post instead of looking for excuses to dismiss it.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 6 months ago (1 child)

    You write like an LLM dude. You use em dashes, you are overly verbose and you use common techniques like "it's not x, it's y" over and over.

    If you're not an LLM you're desperately trying to appear as one. I maintain that you're a bot and I'll happily mute your account now.

  • source
  • parent
  • hideshow 1 child comment
  • [–] [S] 0 points 6 months ago

    Calling every post you dislike AI is paranoia, not insight. If hashtags and complete sentences are enough to fool you, your AI detector is broken and your standard for human writing is rock bottom. Engage with the point or keep shadowboxing an imaginary bot farm. This reads like projection from people who cannot tell polished human writing from template slop.

  • source
  • parent
  • [+] -6 points 6 months ago (4 children)

    Mods don't need to over moderate. If an account is producing good content, great.

    If you notice misinformation, flag it.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 4 points 6 months ago (3 children)

    Is AI pandering good content though?

  • source
  • parent
  • hideshow 3 child comments
  • load more comments (-1 replies)
    [–] 1 point 6 months ago

    Community infra can scale. Will it?

  • source
  • [+] -18 points 6 months ago* (8 children)

    Fdroid and the ecosystem are trash. They don't have the manpower.

    Yes they have a good mission and I appreciate and use the work, but it's grade A garbage compared to the ecosystem and functionality.

    Don't conflate the two and accept the reality and get others to help fix it.

  • source
  • hideshow 8 child comments
  • [–] 11 points 6 months ago (2 children)

    Ignore the f-droid app... consider the f-droid idea and check other clients, such as Droid-ify which are more polished. The idea is that f-droid is simplistic, won't use your data and won't show you tons of ads.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 7 points 6 months ago (2 children)

    What makes Fdroid trash exactly? My experience is really positive, so I don't understand your sentiment.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 6 months ago (1 child)

    How are you defining "the ecosystem" in this context?

    F-droid is not "glossy commercial", but in general that's why it's great.

    I agree it's clunky, and could use a facelift, and a good UI redesign. I would never call it trash though. It's far better than Play or App Store in terms of protecting the user.

  • source
  • parent
  • hideshow 1 child comment