IIRC they stopped collecting data briefly because it was affecting the sites they were checking. They must have fixed that since it’s running again

The description sounds more like an AI receptionist than an AI nurse. It would be helpful if patients could ask follow-up questions to the automated phone call before an appointment. Some clinics don’t have the manpower for that, and especially not in all the languages that the local population might speak.

I’d be interested in seeing how good the model actually is, and how it determines when to pass it along to a human

The concern is with making sure the AI model is only used where it makes sense. Those who are looking to cut costs will try and use it everywhere, and that needs to be kept in check

The GoodKarmaToolkit in particular is an extra project that is managed by ArchiveBox, but the listed services aren’t made by them. I’m not as familiar with ArchiveBox itself, and it looks like there’s an open issue about AI stuff: https://github.com/ArchiveBox/ArchiveBox/issues/1139

There is another called Data Hoarder that does this job too.

Yup, Hoarder was the one I was planning to use for bookmark management: https://hoarder.app/

[no major spoilers in this comment]

I asked two separate friends in the medical field for what they thought of the show, and both of them started with saying that the character is annoying 😄

I think that’s the goal, since there ARE people like that in the field that are difficult to work with. It also gives her room to grow and be ‘humbled’ through the course of the show.

The key is that you need to subscribe to the type of content you want to see. There’s no company here deciding what your feed is going to be.

@BackgrndNoize@lemmy.world, what kind of stuff are you interested in? I’m happy to help you find related communities :)

I can also see some people being opposed to them spending the donation money on ads, since they’d be giving money to companies that may be in opposition to what we’re doing here (or ideological reasons around the advertising industry in general).

Maybe if there was a separate pool of donations specifically for advertising, then people who want to support that can donate to it? Those who don’t can still donate to the projects themselves

Yup, the installation guide says to comment out the things you don’t want to run

There’s the good-karma-kit, which is a Docker compose bundle of some popular projects: https://github.com/ArchiveBox/good-karma-kit

It could act as a list to go off of, if you don’t want to host all of them. The link has more info on each, as well as which ones are non-profit / for-profit

Overview

Have some space computing power and want to donate it to a good cause? How about 10+ good causes at once?

♻️ put an under-utilized system to good use
🚲 use as much or as little CPU/RAM/DISK as you want
✨ 100% more soul warming than mining
📈 geek out over your CPU/disk/bandwidth stats on the leaderboards

This is a collection of containers that all contribute to public-good projects:

• networks: Tor, i2p
• computing: boinc, foldingathome
• archiving: archivewarrior, zimfarm, kiwix, archivebox, pywb
• storage: ipfs, storj, sia, transmission

This v1 list was started by the ArchiveBox project, but it’s open to contributions.

!tech@programming.dev is decent if you’re looking for the more technological side, since the rules filter out

• Minor app updates
• Government legislation
• Company news
• Opinion pieces

Users are concerned that this moderation tactic could be abused or just improperly implemented.

This is the key bit. It’s good to try and make safer online spaces. But Reddit’s automated moderation has been bad for a while, and this might get more users caught up in false positives

I’ve seen comments tagged as abusive regardless of the context:

• someone quoting a news article
• someone making a hyperbolic joke (especially in gen-Z subs)
• actual abuse

For well moderated subs, the vast majority of those reports became false positives over time. For the mod queue, this didn’t affect the end user since mods can dismiss the false positives. But automated ‘scores’ won’t account for that.

We’re going to see even more annoying algospeak like “unalive”, only it’s going to be in news quotes as well

I think it’s a joke about dead internet theory, rather than userbase size

https://en.m.wikipedia.org/wiki/Dead_Internet_theory

The joke comes from an increase in bot use on Reddit, and the subsequent false positive / false negatives in trying to figure out which ones are bots

Lemmy has that problem too, but it’s much smaller in scope. Mostly because there’s less of a reason to try and control the narrative on this smaller platform, but also because the goals are different. Lemmy instances get no benefit from a bunch of fake engagement, and public upvotes makes it easier to catch manipulation

Cross posted to !opensource@programming.dev, and looks like someone already shared it on !thunderbird@lemmy.world

math requires it, the risk is just too high

Interestingly

https://lemmy.ca/comment/5057563

The original reddit was closer to hackernews than the generic site it is now. Not only that but spez has admitted that the original traffic was artificial, by which I mean, the owners themselves were creating fake engagement through various means, such as scraping and cross-posting content from sites like digg via sockpuppets to appear that the site had way more traffic than it had.

Yup, and Reddit started in 2005 (which is 2 decades ago now) with its large migration in 2010. Lemmy only really got going in 2023, and it’s growing

Misskey is a Mastodon style platform, that is popular in Japan and existed from a while back. They added activitypub support in 2018

https://en.m.wikipedia.org/wiki/Misskey

I usually don’t check them unless I need to reference something, but I recommend the alternative UIs for looking through old messages. The default Lemmy UI makes it annoying to follow along in past conversations, while Photon for example has a chat layout

I think the important part is about who is running the server, rather than who made the software

The fediverse is interesting in that context because each instance can decide where they set up the infrastructure or how they process data / requests. The same applies to self hosting

I saw an article that outlined which country each fediverse platform “originated” from, such as Canada for Pixelfed and Germany for Mastodon. That’s fun to know about, but otherwise not important to users compared to the instances themselves

At most it might speak to which laws will govern the project itself, but even then someone can fork a project that goes astray

I think this is being worked on / discussed in the background for how to best implement this. There are a lot of complications when you get into the details of how it should work

In the meantime, another perspective is that even on centralized platforms there will be multiple communities for a particular theme. Especially at first, there may be a few communities for something before one of them will win out as THE community for it.

That’s already happened here for a few communities, where there is ONE main community for a topic.

For others, it’s still in the early stages or (like on Reddit) there are multiple concurrent communities for ideological reasons.

The exception would be if something is created or thought of on the platform, in which case there may only be one community for it from the start. For example, !taneggs@lemmy.ca was a fun idea that started here and has grown into a solid community now (thanks to @DaMonsterKnees@lemmy.world regular posts 😊).

I have only tried Zen from your list and it’s been nice so far. The most recent update last night broke something with the multi account containers, but other than that it’s been smooth sailing for months.

Ladybird looks promising but it’s not out yet. Planning to try switching to it when it’s out.

Arc is apparently dead (or dying), but it was chromium based, VC funded, and Zen does most of the same things anyway. https://www.theverge.com/2024/10/24/24279020/browser-company-ai-browser-arc

The UI mostly, and because the layout is familiar to them