I made this website to crawl and display Lemmy instances since the existing ones were lacking certain filtering and search options.
I have licensed it as MIT and the code is on https://github.com/tgxn/lemmy-explorer
Let me know if there’s any issues, and create a GitHub issue if you want some features :D
You must log in or register to comment.
Damn, this is real impressive! Nice work!
This is one for the bookmarks! Amazing!
Very slick, good work!
What’s the reasoning for “Hide no banner”? There are lots of great communities without a banner, and having that filter checked by default will make those communities invisible for most users.
Edit: one more suggestion would be to include uptime % in the instances list - that’s actually a super important detail for users looking for a home instance.
Thanks for the feedback! The reasoning for the banners thing was that the best communities should have one set, and there are a lot of communities, and the display looks better with them 😁
There was another comment on this, so I’m just deploying a change that will have it default to not hide them.
I could also definitely improve the layout for communities without a banner and instead show some additional details about them (longer description, languages, etc)
Regarding uptime, I don’t really want to have to monitor every instance, but as I’m looking to build a “Join an instance” page I will want to have this data, I may look to use a similar method to this one which uses https://api.fediverse.observer/
Now if there was a way to subscribed directly from the results, I’d say this should be core functionality. Amazing work 👏!
I have a backlog item for that feature 😉 I think it should be possible with the Lemmy websocket API! https://github.com/tgxn/lemmy-explorer/issues/5
This whole page should be either integrated into lemmy, or at the very least linked to by default (in the search or communities page). Discoverability of communities is severely lacking.
This is way better than what else is out there
How are new communities added?
I run the crawler portion of the script - I’m looking to run it in a container and have it update every day.
I put a little thing in the top-right menu that shows you when it was last updated (and only data that was updated in the last 24 hours is included).
This is great work, thank you very much! I have already found a couple communities that I couldn’t see before.
Keep up the good work!!!
Very cool, thanks for your time and effort!
This is extremely well put-together. Much more useable than browse.feddit.de. Thank you for this! It should definitely be included in any guides for getting started with Lemmy.
Looks great! Is there some filter on what communities show up in the search? I noticed that no communities from lemmy.ml show up in the list.
They should all show up. :) lemmy.ml was probably under some load when i crawled it last. I added it manually and they should show up now.
You could add another row of numbers for Average Monthly Users, Number of Communities, and the software version of an instance (Many are on old versions that lack features and security and bug fixes)
Yeah I’m working on another couple pages that will show a better data-table of instances. With details like Incoming/Outgoing Blocks, Uptime and stuff like that.
I’m not saying that the table is a bad idea, but anything in that table and not on the main page will be ignored by most people that visit.
It’s a good point - I think it’s getting a bit cramped, maybe I could roll-up multiple metrics into a “score”, like if instance version is <3 out of date, it gives a lower score.
I’m one of the few that would be interested in the table.
I like tables too :) Maybe just a “change view” button on the list pages.
Why is “hide no banner” default? There are some huge communities without a banner image. Kinda silly.
I’ve just deployed a change with the default to not hide them.
I’m also going to look to improve the display of communities that lack a banner.
Great work! Actually wrote up a quick-and-dirty script to find and list top communities for me but this is quite sexy I’m gonna use this instead
You can also run the crawler yourself with a local Redis instance, and then analyze the data however you want. :)
Definitely bookmarking this. 😁 Thank you so much for creating this!
