Yes its a bug in the crawler, it throws an error about too many files. The problem seems to be due to recursion, because there are up to 2700 concurrent active tasks. Need to find a way to optimize this.

Of course you are free to create such a website.

Tools written in Rust are generally both performant and correct (few runtime crashes or other programming bugs). There arent many other languages with this combination of advantages.

Can you open an issue for this? Its a tricky issue because it only happens randomly, without a clear way to reproduce it. So if you have any useful information like steps to reproduce or logs (browser console and server) then post that on Github as well.

Does anyone have access and can share the paper here? Specifically I would like to know how this can be built or bought, how it works and how accurate it really is. The article mentions an operating live of six months which seems very short, depending on the cost.

You can just take the smallest vps available and it will be sufficient, unless your instance is very large. For distributions Ubuntu or Debian should be easiest.

What do you mean by a “regular fedi post”? Something like a Mastodon toot made outside of any community? It’s not possible to view that from Lemmy, because Lemmy fundamentally only supports content which is in a community (except private messages).

There is a repost button when you view a post in Lemmy (icon with two squares).

Yes this is no problem. Lemmy-ui connects to the Lemmy backend over websocket api. As long as both have the same version they will work together. For example official Lemmy backend installed via Docker, combined with ith manually installed lemmy-ui directly from git repository. You can apply RTL changes to lemmy-ui without any problems as long as you don’t change the api code.

In general backend and frontend should also be compatible if they only differ in patch version (eg 0.17.0 backend with 0.17.1 frontend). But this is not being tested so there is a higher chance that things break.

We live in a fascinating world.

It sounds like people who are commenting there dont know about Lemmy, or about the Group federation FEP.

There may be settings, but most users go with the default which means public posting.

Right that should help, but most content on Lemmy or Mastodon is completely public.

When have scammers ever cared if something is legal? And corporations like meta can pay a lot more lawyers than you.

Most Fediverse data is public so its very easy to scrape. Facebook wouldnt even have to implement any federation in their own platforms if thats their only goal.

Everything in this thread is locally on lemmy.ml, so theres no federation involved ;)

If you have enough time on your hands you could create a blog theme for Lemmy. I believe all the necessary functionality is there, it only needs to be rendered differently.

Reports federate between Lemmy instances. I dont know about Mastodon, someone would have to test and report if there are problems.

This has been fixed already, otherwise you would see HTML in the title of this post as well.

The language should be chosen automatically if possible: For posts: - If the community only allows one language, it should select that - If the user only has one language, it should be selected by default - Otherwise, if the overlap of user languages and community languages contains one item, that should be selected by default

• For comments it should automatically select the language of the parent post/comment

So communities should definitely set their allowed languages to help automate this. Instances as well, as this will also limit community languages to be a subset of instance languages.

Its possible that there are still some bugs. I also think that for the default language selection when posting, “undetermined” should not be considered

cc @dessalines@lemmy.ml

Posts would get federated once they are approved by an admin.

For that it might be more effective to sort by new comments on the main page.

I am working on a feature which might help with that. What do you think about it?

https://github.com/LemmyNet/lemmy/issues/2601

I never suggested that it would be sufficient. Did you also read the rest of my comment?

Yes and there is nothing wrong with talking about these people. Sure the post is not ideally formulated but thats hardly a reason to shut down an interesting discussion entirely.

It’s one thing to not criminalise a couple of horny 14 year olds, it’s another to create a system that sanctions adults sexually abusing 9 year olds.

Now you are completely moving the goalposts, your previous comment was specifically against legalising underage sex. And no one here suggested that adults should be allowed to sexually abuse children.

I dont think you are discussing in good faith here.

To attempt answering your question, I dont think its solely a matter of sexual development of the body. Mental development is much more important, especially the ability to manage human relationships and set boundaries.

Age of consent is also not simply a binary question of yes or no. One approach I know about is that relationships between young teenagers are only legal up to an age difference of 2 years or so, not with adults. And the older relationship partner holds much higher legal responsibility if any abuse happens. I think that makes a lot of sense. It also makes sense that parents should be able to decide if their child can enter in a relationship with a specific person, after all they know their child best.

How do you come to that conclusion? Age of consent is 14-16 in most European countries, and there are none of the problems you mention.

https://www.ageofconsent.net/continent/europe

One Mastodon maintainer participated in the discussion, so at least they are aware of it.

It took some time to deploy, now its working.

If you search for sshfs you can find lots of different guides, like the one below. Basically you need a normal ssh login for another server, and use that with sshfs command to mount a remote folder to the pictrs folder on your Lemmy server.

https://www.digitalocean.com/community/tutorials/how-to-use-sshfs-to-mount-remote-file-systems-over-ssh

If this setup is slow then you can setup caching in nginx for image files on your fast server SSD. That way only a fixed amount of storage will be used to store frequently loaded images. Only images which are older or viewed less frequently will be slower to load as they need to be fetched from the remote server.

https://docs.nginx.com/nginx/admin-guide/content-cache/content-caching/

Right, I updated the crawler now.

I see your problem but I dont think this can be fixed with any rule change or Lemmy feature. It would be possible to let people access posts after they are removed by a mod, but that wont help if your account gets banned. Or if your account gets hacked and deleted. Or if the instance goes down permanently for some reason.

If you are worried about your content disappearing, you should keep backups. For example with an API client which regularly downloads everything to a local file. There is also a feature request for a functionality to export an archive with user data. Even better would be an external service like reveddit.com which reads content from the API and stores it.

I suggest you create a new post to discuss this problem, then more people can give their ideas and opinions.

The problem is that languages on your instance’s version of /c/announcements are also limited to English and German for some reason. I thought this might be a bug in that the site languages would also apply to remote communities, but I cant reproduce that. One thing you can check is go into the db, and make sure that select * from local_site; returns exactly one column. Then take the value of site_id and run select * from site where id = **site_id**;. This should show the domain and other data of your instance.

Anyway you should be able to solve it by refetching the community to your instance (by pasting the community url in the search field). If it worked then curl "https://lemmy.schuerz.at/api/v3/community?id=46" | jq.discussion_languages should output an empty array.

Dont worry, we definitely appreciate your effort :)

Its also a UI issue as you shouldnt be able to distinguish comments in remote communities as admin.

Thanks for reporting!

https://github.com/LemmyNet/lemmy/pull/2697

Did you change the language settings on your instance in between?

Cant reproduce this. Whats the exact error message?

Yes with Activitypub federation you generally have to trust server admins with the data they handle (similar with private messages). Though I suppose it would be possible for communities to anonymize votes when announcing them to followers, then only the admin of the community’s instance could see who voted.