Question sparked after seeing the post about unmarked llm bots.
Would you like to see service/utility bots here?
You must log in or register to comment.
No thanks, those were annoying and spammy on reddit and I think they’re generally a nuisance.
Lemmy let’s you disable bots. Amazing feature.
Hey that’s really interesting and I don’t know how to access that. I access lemmy 99.9% through the voyager client, do you know if I can access that option through the client or do I have to log in on a browser?
In Boost, its under “account preferences” where you can also alter your nio. Not sure if its the same in Voyager
Hmm, I can’t seem to find it in the options. Typically you can go to the website, change the setting, and it generally flows through to what you see in apps. It can vary by setting, though.
I like the useful bots that summarise stuff and you can decide to read the whole article. The “funny” bots can go die in a pit.
Upvoted the legit question.
The only bot I would like would be an improved version of https://schedule.lemmings.world/
Currently, you can scheduled pinned post on a defined frequency, but you cannot unpin them automatically (e.g. after 24 hours)
Some bots that I do (or would) consider useful:
- CommunityLinkFixer
- A configurable AutoMod for communities
- Decronym
- RemindMe
- MetricConverter
- Stabbot
- MoreJpegAuto
- Bridging bots to interact with Reddit
- Bridging bots to interact with Facebook Groups
- Match threader bots for Football, Basketball, and American Football
- Bots that follow tags on Mastodon and boots them to specific communities
If the summarising bot worked better and made short summaries, it would be great. There are too many posts that are just a link with a generic title, so a brief summary would be ideal.
I don’t think there are any other useful bots though.
A moderation bot that pings (DMs) moderators if something goes wrong. For example on spam, or if a comment gets massively downvoted. Maybe it could also detect brigading and other malicious behaviour. Or do semtiment analysis and point out hateful comments. Or scan images or enforce specific community policies.
I spoke with the instance owner of lemmings.world in matrix before and he invited me to try out his webhook system for Lemmy, which is extremely useful. I’ll try making something with that. Maybe an automatic system where the bot moderates every community that it gets added as a mod to.
Awesome. Once it becomes useful, make sure to make it open source and link it somewhere in one of the Lemmy or Fediverse meta communities.
Yeah. Though I always open source my projects, no need for it to be useful for me to open source it.
If you’re interested, here’s the webhook system: https://github.com/RikudouSage/LemmyWebhook
Bot is open source https://github.com/Demigodrick/community_bot
It is, I just asked the creator how to use it on other instances, as the ReadMe states
Note - this is the Lemmy.zip version. For a version for other Instances to run please use (Lemmy Mod Bot) ##link to go here##
RUNNING THIS VERSION WILL NOT WORK ON YOUR INSTANCE WITHOUT CHANGES.
the only bot I’ve ever found useful was on discord, and it’s a bot that lets people coordinate their schedules. other than that, every bot I ever encountered on Reddit was just annoying
There are two separate takes in this thread, those that like bots and those that find the posts spammy.
What about having one bot that provides multiple functions? All the “utilities” can be included in one comment, reducing spam.
Assuming the bot is open sourced, people can contribute modules to it. After that, an instance can choose to run the bot with modules that it thinks are helpful and users can give feedback on what they like / don’t like.
On the other hand, having separate bots allow people to filter features they want or not
Good point yes :)
