I would like to see recommendations for communities based on my communities. It’s not trivial to solve, but discoverability isn’t great right now.
Problem here is also that your instance may not know about all communities from the instances you’re connected to. This could probably also be improved.
Yes, that’s what I mean by not trivial, a centralized system can do analysis like this a lot easier. But even on your own instance, they could find the N users with the most overlapping subscriptions and check which communities they follow to give you recommendations.
I guess the ‘simple’ way of doing this would be adding tags to communities like ‘art’ ‘hobbies’ ‘sport’ ‘football’ etc. This might then let the app suggest others based on the tags you are subscribed to.
It would probably still require some AI/analytics to work out the links based on user activity in different communities/tags but I think it would make it easier to group interests and promote smaller communities.
It could also improve Lemmy visibility in Masterdon if the tags are used as hashtags or something. (Would require more work)
Kbin lists “related magazines” which are similar communities in the fediverse. Not sure how it works but I think it may be based on hashtags like this.
Exactly this. I like how substack works and have no idea if it’s doable or not.
Hashtags could possibly help with this. When making a post, a user can add hashtags which categorize the content. One can then search hashtags, or subscribe to them to find new communities. Probably not as passive as you’d be looking for, though.
Recommendation algorithms are a big reason for the enshitification of other social media. You don’t need to be connected to everything everywhere all at once. Enjoy your handful of small communities.
I don’t want random posts to appear in my feed from communities I haven’t subscribed to, but I want to have a feature that shows me suggestions for communities when I ask for it. That’s a big difference. Right now it’s (too) hard to find these communities.
Recommendation algorithms are fine as long as they’ce user-centric and opt-in.
Split NSFW into NSFW and NSFL.
More options around that in general. I would love a spoiler flag that does the same blur as NSFW but isn’t filtered out by the ‘show NSFW’ checkbox.
It would probably be better to have a more general tag system and then NSFW and NSFL could just be examples of tags.
Although NSFW really serves the extra purpose of “18+” which is important to have for legal reasons.
Maybe a setting for each tag for whether it qualifies as NSFW? That way you could have multiple tags that would be filtered as NSFW for different classes of content, which could enable individual users to only filter one of the tags if they only want to avoid something specific.
I’ve made a post a few days ago. I’d argue we should make a proper distinction. Adult content and NSFW isn’t the same thing. Currently everything from sex education to gore and death is the same category. I think it’s really not. NSFW tags help so you can scroll through things in an open-plan office or while commuting. Porn is porn and gore is gore. I think we shouldn’t oversimplify this but keep the nuances and have different categories. Also I’d like to not mix stuff like sex education which might be fine, and minors ask those questions all the time on Reddit with other things like fetish.
An option to view all comments from crossposts when browsing a post. It’s annoying how you can see a post that’s been crossposted 5 times and wonder where the comments are.
Reading some of these is making me really appreciate @ernest.
We have this one, it’s handy. There’s a list of crossposts and how many comments each has, you can click to where the acive discussions are.
Oh hey, I had no idea ernest was back. Great to have him back.
Yeah we heard from him recently. Hopefully his health is good now.
Cross instance post/comment deletion.
Sometimes I just don’t want my comments to live forever and deleting shouldn’t be impossible.
I don’t know, man, your comments are works of art.
Nah, not mine, piss takes and sarcasm mostly.
Exactly. [chef’s kiss]
The fediverse is not set up for that. Everything is public and could be saved by an instance admin
Sometimes I just don’t want my comments to live forever and deleting shouldn’t be impossible.
Due to the nature of the Fediverse, this will be very difficult if not impossible to implement/enforce.
detection
You meant deletion? This is already implemented about as well as it can be.
deleted by creator
-
Reports categories based on both the community, the instance of the community + the user to reduce report noise between mod actions and admin actions.
-
Post tags, to label content within a community.
-
Better language support, clearly indicating which ones are allowed when submitting something in the language dropdown, as well as basic language detection support.
-
When the instance is using pictrs, add a section in the user’s settings to see all the uploaded pictures in that account, with the ability to delete any of them.
-
Better accessibility / a11y support for uploaded images with alt-text.
-
Support for svg-based emojis
-
For mods, the ability to make a pinned post made by one of the mods editable by other mods, which would be useful for FAQs, etc.
-
The ability to subscribe/follow a specific user, not just communities.
-
Passkeys support as a 2FA method.
-
Some basic builtin automod action, such as blocking known keywords from spammers from being posted, not just showing as removed as when using the slur filter in the admin settings.
EDIT: Something I just thought of
- A URI protocol handler to refer to communities, users, post and comments in an instance-independant way (ie:
lemmy://u/[email protected],lemmy://c/[email protected],lemmy://c/[email protected]/p/1234567) or another syntax that makes more sense. That way you could let the OS redirect the query to the software of your choice, and define your home instance there.
Now there are some issues to figure out before defining the URI handler, like how to refer to a post or comment that will redirect to the appropriate one on your home instance since post and comment currently have a unique ID on each instance, which makes them hard to directly address without doing some kind of conversion.
Tags for posts would be great!
-
The ability to see all the communities if you search them up without having to find it via lemmyverse.net and inserting the specific fedi url to the search bar. It’s a crucial thing for an average Joe, no matter it’s due to how the fedi protocol works.
Hmm I only browse in mobile voyager and there’s an “all” filter. Is that not available?
That ‘all’ is all of the communities and posts your server knows about. You are on a pretty big (I think?) server, so it’s probably pretty good. For people on smaller servers like the one I’m on, it won’t have a lot of the smaller niche communities on there as no one from my server has ever visited them.
If I made a new community on my instance and posted stuff there, you wouldn’t see it in your ‘all’ feed unless someone from lemme.ee visited the new community first.
Nice explanation
Someone made a tool for you to automatically add a community to various all feeds from instances that are part of the service. I’ll try and find it.
Honestly I feel like Lenny needs flairs more than anything
I agree. It’s one of the number 1 features I miss from Reddit.
Guessing fixing child porn propagation isn’t the highest priority?
Make it easier for server admins to connect/link to the child porn hash databases, scripts for autobans + deletion of any content, flagging + notify to other servers etc.
This should actually be #1. Last thing I want for the fediverse is for it to become a CP haven because we lack proper moderation. I recently received an amazing presentation on the issue of CP distribution from a seasoned officer, and CP is a genuine and dangerous issue. Go look up what sextortion is.
Allow communities that contain the same content but exist on different instances to show each others content as if they were one community.
That sounds like it’d be fantastic for reading but, depending on how it’s implemented, hell for posting.
Lemmy already aggregates posts from communities you follow into one feed. If it allowed the creation of an arbitrary number of sub-feeds configurable by the user, that would be incredible. But every user would have to build these on their own from scratch. Great for user choice, but no communities will come bundled by default, so small communities won’t get a discovery boost.
If instead there was some kind of first-class notion of a “supercommunity” offered on the server side, where it acted as a transparent view of other communities, that’d be a great visibility boost for small communities. But if you tried to post to it, which underlying community would it post to? You’d have to either designate a default community to receive posts (which would be unfair to every other community there), randomize where it goes to (which would be a quagmire, what if your post is allowed in half of the communities present but rule-breaking in the others?), burden the user with choosing (which would be hell if there are a lot), or simply make it read-only. I don’t really like any of these. It also raises hairy questions about who will control which communities are and are not part of the group, how the groupings react to defeds, etc.
deleted by creator
Like multireddits on that other site
Except multireddit were not shared accross users, making them largely irrelevant.
On lemmy, the default view should be, if you go to /c/books, you get all books on all instances in a single place.
Anything less will suffer the same downfall as reddit
What if someone created a /c/books on their own instance with bad intentions, and filled it with propaganda, porn, and ads?
Just block it We can share block list, preferably automatic consensus blocklists and gradual deamplification list. The alternative is the current default which as if they were all already blocked, except the one big one.
Yes, until that’s the default, lemmy is destined to repeat the mistake of reddit
There NEEDS to be an account migration option, with not only settings but also my saved posts and comments, own posts and comments etc. If not possible, at least allow an export in the style of a gddpr dump from the likes of facebook etc. to allow import at a later time when implemented.
My instance is shutting down at the end of the month (~500 users) and there is no good way to export my data. I would not be surprised if some of the 500 get frustrated and stop with lemmy.
Migrating posts and comments is not possible with activitypub, as that would be rewriting history. But you could open up an issue for a user data only export, as that wouldn’t be too difficult to do.
Thanks for your reply! I looked into it and #3976 seems to be pretty much that because the “import” part of it was shot down.
I could create an issue for specifically gdpr style exports tho.
That’d be good, thx.
Speaking of which, a mechanism to manage posts from instances your account has previously migrated away from.
You mean like Mastodon ?
Also not resolved https://github.com/mastodon/mastodon/issues/12423
Moderation tools. They need to drop literally everything else they are working on and build robust moderation tools for community owners. Nothing else matters more than this.
Could you link to the issues in the lemmy issue tracker, so I can know which ones you’re referring to?
Here is a hopefully minor thing…
Reddit has multireddits where you can have a few that follows a certain selection of subreddits under a label. You can have multiple ones defined as well. Therefore, you can have a view for all things news (following multiple news things) without having to view those things on your main home feed (as well as any other defined topics that you can think of).
It would be nifty if such a thing could exist inside of Lemmy as well.
We have something like this at Kbin, including the ability to make them public and subscribe to other peoples’ public multis.
It’s really great, I hope Lemmy gets it.
Edit: here is one of mine: https://kbin.social/c/Cinema
There is this issue for it https://github.com/LemmyNet/lemmy/issues/818
For now at least, if you use Sync you can have multiple accounts each with their own subscriptions and settings and switch between them at will.
I’d like to see more instances with 100-500 users.
I know that’s a community thing, more than a Lemmy thing. I just don’t feel like I have a wealth of choices. I’m still on lemmy.world and when I look around, I don’t see a lot of medium-sized instances to migrate to.
Try to join an instance that is related to your geographical location or your country or state. That should result in a more even spread than what we have right now.
I’d love to see more country specific instances.
I’ve looked. I find it hard to believe that there’s no California or SF Bay instances, but I haven’t found anything.
“be the change you want to see” and all that. Just takes 1 admin to take the initiative.
It’s a lot of work to admin an instance, though. I can understand if random people don’t want to do it. I hope one day someone does, because a California or Socal instance was the first one I looked for, too.
In-line translation features for non-English communities (in my case) would be very helpful and would exceed Reddit functionality, which is something I think Lemmy should strive toward
Integration with DeepL API like Misskey would be cool!
While we are speaking of it, please let users choose between languages (original and target) independently of system locale.
Sometimes I encounter social media posts auto translated (probably through Google translate) but languages detection is messed up. And the best part is there isn’t a menu to choose.
Yes this would be amazing!
This should really be implemented at either the browser or lemmy-ui/app level, not in the back end.
Brigaders.
People with like 10 accounts that upvote themselves/downvote dissent.
Kbin votes are public so we can see brigades in action.
Apparently Lemmy votes are too but it’s not accessible in the native interface, only from Kbin. Maybe a third party client will implement it.
It’s super obvious when it happens to you, but it’s not obvious when you see it in the wild. It would be a great improvement to the site to just show the users who downvoted/upvoted.
Admins and mods can see this in the latest version.
It really is. I’ve seen people being called out for doing it on kbin because we can literally see a list of users who upvoted.
Strong disagree, I think it has a chilling effect on voting. I think there are plenty of reasons why someone would want to lurk without attaching their account to a given topic, and this is inevitably going to cause problems with targeted harassment down the line as Lemmy grows.
That’s a double edge sword though. I’ve already had unrelated votes thrown in my face a couple of times. Honestly I really wish instances could choose to anonymize votes if they wanted.
That’s annoying. “Double edged sword” basically sums up social media.
In other discussions I’ve seen people say that non anonymous votes are baked into activity pub.
It is actually interesting you can see that comment. I am apparently banned from .ml again. For making a joke about .ml bans in a .world thread.
This is exactly what I’m talking about though. I have clearly been marked by .ml mods and now that seems to be following me across the fediverse. You can look at my profile and see that there’s nothing really controversial there, I’m not a troll, and don’t flame or antagonize people. The only thing I can come up with is thay I’ve been marked for voting patterns.
Maybe I’m being paranoid, but I can’t help but think the vote functionality was designed intentionally to enable exactly this kind of wider fediverse censorship.
As far as I can see from the mod logs, your last ml bans expired 2 days ago (and were probably due to saying negative things about a couple of specific countries).
Blocking is strange though and doesn’t really censor well. There’s a guy from kbin who likes to vociferously troll in ml and has been perma banned from it but everyone in kbin can still see and interact with his posts when we go into ml threads, much to everyone’s irritation.
I think the fediverse’s strength is that it’s decentralized. Blocking and defederating doesn’t silence people, it just changes who can see them from where.
You have to actually go look at the .ml mod logs specifically. That’s another thing which is quite weird and suspicious. Certain .ml actions don’t seem to appear in the federated modlog. This example is particularly spooky because it was via a .world thread.
It kind of seems like a bug tbh. Like they have a tool to sanitize the mod logs, but it doesn’t change the local copy for whatever reason. IDK. Honestly the shady shit on .ml has really turned me off of fediverse to a large degree. I lasted on Reddit for 15 years without getting any bans, and now this supposed post-censorship community has me marked for reasons I can’t quite figure out and nobody will explain.
I did go and look at them on their instance though (I’m on Kbin so doing that is the best way to see that stuff). I searched by your name both in this sub (no bans) and on the instance (expired bans, but I didn’t exactly go through it with a fine tooth comb). I saw the screenshotted.
Like I said I can see you here right now, just fine. The lemmy devs are the admin mods on the ml instance and it’s well known they have certain political rules. They are marxist leninist. If you’re repeatedly catching bans from them it’s probably best to concentrate more on other big instances that don’t come with that.
That’s a problem not a solution.
Literally my first experience on Lemmy was arguing with a troll who was getting 3 upvotes within the first second of posting, while handing out three downvotes instantly as well. It was obvious, and when I called them out they said “I just have some fans.”
