One thing I’ve noticed is the bot seems to “correct” people who include both the Lemmy-friendly link and the actual URL in their post. I’ve seen this a few times, although it’s possible the posters are editing the posts to add this in response to the bot. It’d be nice if the bot could check for this.
Fyi your bot is apparently trying to “correct” people posting from Mastodon too. Seems like it’s replying to every comment that comes through from Mastodon’s side. Example: https://lemmy.world/post/1628806
Yes, I know about that and I think it’s working as intended, only the message sucks. I already changed it today, but now that I’m looking at it again, it still sucks :/
Do you have any idea for a better message? There are two messages currently, one for Lemmy users and one for non-Lemmy users:
Lemmy users message: Hi there! Looks like you linked to a Lemmy community using a URL instead of its name, which doesn’t work well for people on different instances. Try fixing it like this:
Non-Lemmy users message: Hi there! The links in your text lead away from the user’s instance, here are fixed links that stay on each user’s instance:
Well to post to Lemmy from Mastodon, the Mastodon user has to tag the community. That tag displays as a link on the Lemmy side. The Mastodon user has no control over the format of said link, as it’s a Mastodon user tag. So your bot shouldn’t be trying to correct it at all in this case.
It’s just not needed though, is the thing. These “links” aren’t links that are posted for people to follow, they’re just an artifact of having posted from Mastodon. As we get more interaction between the two platforms, your bot as currently written will reply to every single post and comment made from Mastodon, ever. It serves literally no purpose in this situation other than to clog up threads.
I’m not trying to start a fight or anything. If you absolutely won’t be convinced that it’s pointless replying to Mastodon tags, maybe you can at least consider limiting it to replying only on the main post? Otherwise, again, every single comment in the thread that is made from Mastodon is going to get the same bot response.
Not sure if you’re aware but this is still happening. In fact, looking at the bot’s posts today is it possible you accidentally set it to only respond to Mastodon threads? 😅
If you block it, it won’t reply to you. If you ban it on instance level, it will be blocked on all federated instances on all communities. If you defederate, you will block real people on the instance the bot is on.
If you block it, it won’t reply to you. If you ban it on instance level, it will be blocked on all federated instances on all communities.
This is a good point. Yet I feel like there’s a point where the annoyance overshadows the benefits. If there’s an annoying bot on every single instance, is it still reasonable to snipe-ban individual bots? What if a server is mostly bots, yet have real users?
I think it’s better not to accept bots at all.
If you defederate, you will block real people on the instance the bot is on.
This is true. But the argument feels a bit hollow when you plan on blocking Threads. If you are willing to block ten million users, why shouldn’t others block your 40?
Yet I feel like there’s a point where the annoyance overshadows the benefits.
The bot is generally liked, it has much more upvotes than downvotes, so that sounds like a you problem.
I think it’s better not to accept bots at all.
That’s your opinion. But clearly the creators of Lemmy itself think bots should be accepted, otherwise they wouldn’t add a flag to mark user as a bot. And in your settings you can even disable seeing any bots, which sounds like the perfect solution for you.
But the argument feels a bit hollow when you plan on blocking Threads.
That’s whataboutism. That’s an entirely different issue than you disliking bots.
why shouldn’t others block your 40?
Because that’s not how federation should work. Sure, block the instance all you want, but you’re actively going against the spirit of Fediverse, especially when you have many other tools to solve it differently.
there is a bot that corrects the links, maybe we should improve that
Yep, feel free to offer suggestion for improvement (or link to false positives, so I can fix that)!
One thing I’ve noticed is the bot seems to “correct” people who include both the Lemmy-friendly link and the actual URL in their post. I’ve seen this a few times, although it’s possible the posters are editing the posts to add this in response to the bot. It’d be nice if the bot could check for this.
I’ll fix that, thanks!
Thanks for your work on the bot!
Fyi your bot is apparently trying to “correct” people posting from Mastodon too. Seems like it’s replying to every comment that comes through from Mastodon’s side. Example: https://lemmy.world/post/1628806
Yes, I know about that and I think it’s working as intended, only the message sucks. I already changed it today, but now that I’m looking at it again, it still sucks :/
Do you have any idea for a better message? There are two messages currently, one for Lemmy users and one for non-Lemmy users:
Well to post to Lemmy from Mastodon, the Mastodon user has to tag the community. That tag displays as a link on the Lemmy side. The Mastodon user has no control over the format of said link, as it’s a Mastodon user tag. So your bot shouldn’t be trying to correct it at all in this case.
The user may tag multiple communities, I tried to not phrase as a correction but more like “here’s a link for Lemmy users”.
It’s just not needed though, is the thing. These “links” aren’t links that are posted for people to follow, they’re just an artifact of having posted from Mastodon. As we get more interaction between the two platforms, your bot as currently written will reply to every single post and comment made from Mastodon, ever. It serves literally no purpose in this situation other than to clog up threads.
I’m not trying to start a fight or anything. If you absolutely won’t be convinced that it’s pointless replying to Mastodon tags, maybe you can at least consider limiting it to replying only on the main post? Otherwise, again, every single comment in the thread that is made from Mastodon is going to get the same bot response.
Replying to only the top-level post for Mastodon makes sense to me, thanks for the suggestion!
Edit: I temporarily disabled replies to Mastodon posts until I can think of a better solution.
Not sure if you’re aware but this is still happening. In fact, looking at the bot’s posts today is it possible you accidentally set it to only respond to Mastodon threads? 😅
Examples: https://lemm.ee/post/1456125, https://lemm.ee/post/1375340, https://lemm.ee/post/1441558
a
Yeah, banning them on instance level or defederating seems like the worst idea. Blocking them on individual / community level is the way to go.
a
If you block it, it won’t reply to you. If you ban it on instance level, it will be blocked on all federated instances on all communities. If you defederate, you will block real people on the instance the bot is on.
This is a good point. Yet I feel like there’s a point where the annoyance overshadows the benefits. If there’s an annoying bot on every single instance, is it still reasonable to snipe-ban individual bots? What if a server is mostly bots, yet have real users?
I think it’s better not to accept bots at all.
This is true. But the argument feels a bit hollow when you plan on blocking Threads. If you are willing to block ten million users, why shouldn’t others block your 40?
The bot is generally liked, it has much more upvotes than downvotes, so that sounds like a you problem.
That’s your opinion. But clearly the creators of Lemmy itself think bots should be accepted, otherwise they wouldn’t add a flag to mark user as a bot. And in your settings you can even disable seeing any bots, which sounds like the perfect solution for you.
That’s whataboutism. That’s an entirely different issue than you disliking bots.
Because that’s not how federation should work. Sure, block the instance all you want, but you’re actively going against the spirit of Fediverse, especially when you have many other tools to solve it differently.
Oooh is the bot on gh?
Not yet, but it will be (I’m currently rewriting it, the next version will be there). Do you want me to let you know when it’s there?
Sure! I’m mostly a python dev but if it’s js or something I might be able to contribute on a small issue here or there
It’s there now! https://github.com/RikudouSage/LemmyCommunityLinkFixer