Oh, I thought it was about tagging the communities and merging them based on those tags. Are you suggesting tagging the posts instead, and displaying all posts tagged the same from all communities across all federated instances in the same location?
But this can already be accomplished with the search feature. You only need to select ‘Local’ or ‘All’ and search for a word. People shouldn’t be forced to hashtag every post, so the result would be the same as it is now.
I think conflicts can arise with hashtags just as easily as with community names, so it might be better to have an updatable and moderatable link instead.
Lol this is a very good point. I’ll give it some thought.
Cross-instance “multireddits”, that are also automatic and topic-based. #1113
TL;DR: The suggestion is to implement an automatic multireddit feature in Lemmy that displays all posts from communities with the same name across federated instances. It aims to promote decentralization, avoid echo chambers, and ensure high availability. Community moderators would have the option to opt-in or opt-out their communities from being displayed. There are discussions about potential issues such as community name collisions, duplicates, abuse, and practical implementation. Some propose using a new link format, while others suggest providing users with a list of related communities.
Combining communities should be a front end feature… Allow users to merge their views if they want. But it should not be enforced at the backend or federation level.
Eventually there will be third party apps which can do this merging in their interface if someone wants it.
I agree with this. The grouping should be a front-end feature based on hashtags, as someone else mentioned, instead of the community names. Alternatively, there could be lists that you can simply copy and paste to create your own multireddit, eliminating the need for hashtags. However, considering that the original issue was already on the lemmy-ui, I’m not sure why you brought up the backend.
Every moderator would have control only over the content displayed on their instances, and not on everyone else, as it should be. The argument about having one or two large communities is a recurring one. There is no reason to have federation if we are going to centralize communities in a couple of instances. Then, if one of those instances shuts down, everyone in those communities would have to migrate. The main benefit of federation is decentralization.
I don’t know why you are removed about rules and frustration.
I believe the best approach would be to have these multireddits automatically created for convenience. However, users should have the option to choose whether they want to see only the content from their instance’s community or from any number of communities, instead of being forced to view all of them.
I think this would be the best, but it will probably be a choice of each community moderator not the users.
The proposal does not necessarily imply merging all small communities with others. The implementation can provide an optional choice to community moderators, allowing them to decide whether they want their community to be included in the multireddit. This approach respects the autonomy of individual communities and acknowledges the reasons why new but similar communities may emerge, such as issues with community mods or server admins. By offering this flexibility, the feature can cater to the diverse needs and preferences of different communities while still providing the benefits of consolidating posts from communities with similar topics.
The goal of implementing this feature is to leverage the benefits of federation. If we wait until there is only a few big communities, the purpose of having federation becomes irrelevant. When an instance hosting one of those large communities shuts down, the community would have to migrate to the next major community.
By proactively implementing this feature, Lemmy can harness the advantages of federation while actively mitigating the challenges posed by community fragmentation and echo chambers. It provides a centralized hub that encourages cross-pollination of ideas, fosters community engagement, and ensures that valuable content is accessible to all users, regardless of the size or popularity of individual communities.
No. You can hide posts you open with the user setting show read posts on the web-ui.
Here are some of the monthly active user statistics for various social media platforms, according to the search results:
It’s worth noting that these numbers are subject to change and may vary depending on the source. Additionally, there are many other social media platforms with significant user bases that are not listed here. When choosing which social media platforms to use for your brand, it’s important to consider the relevance to your target audience and the quality of engagement on each platform, rather than just the number of users.
Citations:
[1] https://buffer.com/library/social-media-sites/
[2] https://www.semrush.com/blog/most-popular-social-media-platforms/
[3] https://datareportal.com/social-media-users
[4] https://statusbrew.com/insights/social-media-statistics/
[5] https://en.wikipedia.org/wiki/List_of_social_platforms_with_at_least_100_million_active_users
By Perplexity at https://www.perplexity.ai/search/98165c01-5be7-41be-ba6c-fd77454d3b5a
We are at 5k monthly active users. I find that stat more interesting.
Here is an issue that would solve this but ideas are welcome.
I agree with what I just read. It should just be an instance setting and let admins choose. Mastodon can’t outcompete Twitter without an algorithm to easily find things you find interesting, and Lemmy probably won’t be able to outcompete Reddit without karma.
This is one of the features that made social media successful. Without the reward-dopamin loop, people have less incentive to generate quality content. Publicly showing karma is another thing. While it’s nice to be able to hide it, it’s one method of judging a user on reddit. In combination with other methods, it’s easier to spot trolls.
What’s wrong with “karma farming”? Disabling it, obviously doesn’t prevent spam/propaganda/bad quality content. Also, reposting is important for people new to the platform. (Who didn’t laugh at a joke on reddit which everybody seems to already know about?)
I think improving moderation methods would be way better to counter low quality content than radically concealing information that’s already there for the one in control of the instance.
— heeplr
Well, these AIs are being trained on public figures, and there isn’t much they can do unless they livestream with the AI impersonating them, allowing them to potentially identify who is behind it. How will people figure out if there’s an LLM out there that speaks just like them? It’s similar to fine-tuning AIs on artists to create art that mimics their style. It can be frustrating, but there isn’t much anyone can do unless surveillance software is installed on every computer. In summary, I don’t mind because I won’t even find out.
Oh man, it’s going to be so much fun to have to group similar communities when there are hundreds of instances! 🙃
Bullshit. This assumes the people training LLMs are the same ones building the datasets. Once a dataset is created, it can be used to train multiple models, meaning that there’s no further impact on API usage.
Yeah, the fuck? I’ve reloaded and I’m somewhere else than where I wrote the comment.
There could be conflicts between hashtags, as a hashtag for one community may not have the same meaning for another community. This would result in mixing topics and potential confusion.