mke @ mke @programming.dev

Posts

0

Comments

135

Joined

1 yr. ago

About as much as I trust the yanks.

yea this attitude right here is why ai bros are so beloved

Anubis does its thing, shows me cute art, then leaves without elaborating. It's a mostly non-intrusive, individual/community effort to protect people against big tech and abusive scrapers. I usually see it in open source community websites that were getting hammered by LLM scrapers.

Cloudflare's is a corporate solution from the company that man-in-the-middles half the internet and makes me click shit every fucking time. I see it whenever I make the mistake of following a stackoverflow link.

You're goddamn right my reaction is accordingly different.

I think most popular online social spaces—here included—are unconducive to the kind of long-term, active effort conversations between individuals that might change someone's mind. Instead, we gather as many as we can and just start shouting.

Shout it loudly into every medium and hope your fiefdom's propaganda, however meager, pulls in some people over time. It's all about that noise, because noise means engagement, and engagement is the de facto cost-effective tool in social media. Honest conversations are high-cost, dubious reward. They're tiring. Jading. Hardly worth it. So, broadcasting ideas it is.

It's fundamentally the same method the ruling class has used, and continues effectively using, to influence the world in any scale that actually matters, except they reach hundreds of millions every day while we dozen thousand few shout at each other in a dark alley.

Regarding online behavior, you might consider that Lemmy has largely failed to fix it. I posit that it barely ever tried, but also that it's an extremely difficult problem because the root cause is us.

Yeah, the more people say something, the less true it becomes. Good facts die inside an individual mind without ever being shared.

It's not just that. If it was, I'd have been convinced by the new-era freedom fighters using fucking ai horde, or whatever sloppers call their latest open-source [citation needed] model instance.

The biggest issue is the shameless, continuous abuse of creative workers by sucking up all their works without consent.

Turning it against itself my ass. This is promoting it. Legitimizing it. Normalizing its presence everywhere. Doing exactly what they want.

They're on a niche forum sharing convoluted AI slop, harming that which they claim to stand for. These aren't freedom fighters subverting the system, they're clowns.

That's too far, though I understand the feeling.

I think we should save the Hitler comparisons for individuals that actually deserve it. AI bros and genAI promoters are frequently assholes, but not unapologetically fascist genocidal ones.

Another isolated case for the endlessly growing list of positive impacts of the GenAI with no accountability trend. A big shout-out to people promoting and fueling it, excited to see into what pit you lead us next.

This experiment is also nearly worthless because, as proved by the researchers, there's no guarantee the accounts you interact with on Reddit are actual humans. Upvotes are even easier for machines to use, and can be bought for cheap.

The articles mentions that scroll and the arrow keys no longer adjust volume. Nothing could be earth shattering because it's video streaming software, but it does seem to come with some functionality loss at this stage.

It's still nice! A bit of recognition, legitimacy, and although it's not funding, it might be a small step towards it. I see many great works, that stand tall on their own. More eyes will only make them shine even brighter.

Thanks, Fr*nce.

Given the breadth and depth of evidence of him being a horrible person and holding favorable views towards fascism/nazism (see him flirting with every far-right movement), calling it misconstrued is misrepresenting reality. He knows what he did.

Unlike other commenters, I don't think you're necessarily a Nazi, but you are at least missing crucial information to make a proper judgment here. And, not to group you with them, but when you defend his Nazi salute, you're not in good company.

I hope you'll reconsider your stance.

That's nearly every company making image generation AI right now. Stealing from creators is all the rage, and I've yet to see a reasonable defense of this by AI users.

I've seen jokes(?) that they're aiming high, as in, aiming for high global temperatures to handle the ice. Stupidity wouldn't describe it, it'd be insanity... But isn't that what this administration is all about?

It's not perfect, sure, but we as a society should be capable of deciding that some things aren't okay without giving the state carte blanche to censor as they see fit. If the system can be abused, then we ought to fix it, not forgo it entirely.

Plus, governments and companies already suppress or ban a bunch of speech, often in favor of the ruling class. I doubt outlawing harmful speech like parent comment suggests would be the straw that breaks democracy's back.

Most people know this in some capacity, but it's not talked about enough: the shape of the platform massively shapes its culture. Every mechanism, intentional feature or not, is a factor in resulting user behavior and should be accounted for.

Reddit Karma was (shitty) reputation from the start, but Slashdot user IDs became one despite being mere sequential identifiers; negative user feedback such as downvotes can be harmful to communities (yet, users without an outlet may lash out in other ways e.g. reports); even how the platform communicates with users influences them; and so on.

I'm not saying you shouldn't be nice and incentivize others to do the same, but unless the system naturally leads to the desired behavior, you'll have a bad time in the long term because building culture by interactions doesn't scale. By the time you realize there's a shift, it's too late; interactions will compound and affect how the average user acts faster than you can try to course-correct.

I wish lemmy was more experimental, because by building a clone of reddit, we've copied too many of its faults. We've already got gatherings to complain about mods, and the one time devs considered changing a core component, discussion was killed by an onslaught of users. Problems with the current setup that were brought up then will likely never see that amount of people thinking about how to solve them.

Contrast with Mastodon, which gets crap for not being a faithful copy of twitter, but their reasoning for not including quote-reblogs is understandable. They're now putting a lot of thought into how to add them safely. Not ignoring functionality users want, but also not ignoring how it will affect culture, that's compromise.

I'd like it if we could talk more about how our platforms work and, particularly, how they affect us, because that's a big way we can build better platforms, right up there with being nice.

Apparently the dump doesn't include media, though there's ongoing discussion within wikimedia about changing that. It also seems likely to me that AI scrapers don't care about externalizing costs onto others if it might mean a competitive advantage (e.g. most recent data, not having to spend time and resources developing dedicated ingestion systems for specific sites).

I want to stress this: it's not that "tech bros" are just stupid—even though a lot of them are revoltingly unappreciative of the giants whose sholders they stand on—it's that they don't care.

No one who uses Mozilla software wants more cloud shit or online services from Mozilla.

I don't think that's unanimous. I'd like to use Firefox Relay, myself, and I'm willing to give thundermail a chance.

Used to think I'd go full Proton eventually, but leaning more towards a diverse set of service providers, nowadays. It's also my hope that these services allow Mozilla to depend less on companies like Google, and more on the users they ought to serve, which would be healthier for the org and better for users.

Yes, sort of. Thundermail addresses, apparently, or bring your own. From the linked article you're commenting on:

Users can send and receive email using new Thundermail accounts they sign up for. The service will also allow using your own custom domain (e.g. your.name@yourdomain.com).

I should donate again. As someone who still depends on gmail, I keep forgetting how annoying it was to get ads every time I refreshed my inbox, before I switched to their app. Glad things seem to be working out.