It’s easy to discover communities on my instance via the dedicated page in the hamburger menu. But let’s say I want to follow a community on another instance, such as [email protected] . I might have found its name mentioned in a post or comment. When I click on the provided link, I’m thrown on that instances web page, from which I of course can’t subscribe.
So what I instead have to do is to copy the description of the link and paste it in my instance’s search bar. Which isn’t easy, since it’s a link, so there isn’t even a straightforward way to select the link text without clicking the link. This seems very unintuitive and makes the process of joining a whole bunch of communities tedious. Is there a better way?
Would second this. I’m a tech savvy person as I work in IT and even I’m having to think about what I’m doing just to subscribe to different communities, then there’s multiple of the same communities on different instances etc it is quite tedious as you say.
Really struggling to see how this gets mainstream adoption as your average user isn’t going to have much joy… From my brief interaction with the fediverse I think it’s going to become the Linux of social media I.e. for Geeks and Hobbyists rather than your every day user.
Think about everything you hate about Reddit—the kids, the trolls, the spam—and be thankful Lemmy requires a little more effort.
This is the way Reddit used to be when it first came out.
In my experience there are many good and positive casual users on reddit as well as toxic and obnoxious techies. Knowing how to navigate an obtuse UI is not a mark of good character.
No, of course not, but the added…intentionality that it requires weeds a lot out. Remember, trolls usually go after that which requires the least amount of effort. So it’s not about being able to navigate a UI, it’s about effort.
Or maybe that’s just a bunch of bullshit. 4chan is pretty arcane from a UX perspective, and look at the cesspit that place is. I don’t know. It was just a thought experiment.
All of this is true, but I wanted to relate a similar phenomenon that I observed some 30 years ago that might be of interest, or at least entertaining, to everyone here.
In my formative years, I spent a lot of time reading Usenet, which, briefly, is a text-only forum not dissimilar to bulletin boards or subreddits or Lemmy communities.
I frequented one group in particular called
alt.sysadmin.recovery
. Most Usenet groups began withalt.
by archaic convention, and the rest of the name is simply descriptive or categorical. There were groups likealt.hobbies.baking
and so on. Again, not dissimilar from (and likely inspiration for) these modern web-based communities.This group was for system administrators (or “sysadmins”) to generally gripe with one another about the difficulties of their jobs, dealing with users on their systems or networks, and similar. One of the rules of the group was that no advice was ever to be requested, nor given. It was strictly for sysadmins to vent. The key point here is that everyone in the group was in some technical role.
What was unique about
alt.sysadmin.recovery
was that you couldn’t post to it. At least, it seemed that you couldn’t, because the group was set to be moderated, but had no moderators. If you posted a message to a moderated group, the message would be emailed to all of the moderators on record, who would either delete or ignore them, or apply their stamp of approval for the message to be posted in the group.alt.sysadmin.recovery
had no moderator emails configured.The trick is a little bit technical. Usenet posts are quite similar to emails: they have some “header” fields (like the title of the post, its author, and so forth) and a body. Most of the headers are not displayed directly (which is also true for email), such as what Usenet software sent the message, and so on.
When a moderator approved a message in a Usenet group, their client would append an
Approved:
header line with some value, like their name, or the date, or something. As long as theApproved:
header was there and had any value at all, the message would be distributed to the group.So the trick was to simply append that header when you posted the message. Since there were no moderators anyway, nobody could ever accuse you of bypassing the system. Bypassing the moderation system was, in fact, the entire point. You had to know enough about how moderation worked in Usenet to post a message to the group.
One of the lasting results was that
alt.sysadmin.recovery
was never overrun by bots and spam, even as the rest of Usenet became an absolute cesspool through the '80s.Which brings me back to my point. A few hoops to jump through and a few initial challenges to adoption can go a long way as a filter for who can show up and interact. Of course we would want Lemmy to be welcoming to anyone who will make the community better, brighter, more fun, and more useful… But we can take our time cracking open the floodgates. Maybe that’s for the best.
I agree. Lemmy doesn’t have to replace Reddit, it just has to be a working alternative.
Over time Lemmy may gain a much wider audience, but as long as it has enough users to be entertaining and informative, then it will be a good alternative to corporate social media.
This is exactly why I think many comments on Reddit miss the point when they state that “Lemmy will fail because it’s way too complicated for mass adoption”. Maybe not every Reddit participant has to join Lemmy. Maybe it’s good that there is a (small!) hurdle to overcome, that does not exclude or discriminate against anyone, but simply requires a tiny bit of effort.
Or, you know, discriminates against anyone who can’t be bothered to read a short few paragraphs about how something works and follow basic instructions. I’m pretty OK with that particular type of discrimination!
Nice anecdote, I barely used Usenet back then, and I get your point. But I also think that federation is a key element of Lemmy, and it should be made to work as smoothly as possible, for better or for worse. There could always be some obscure communities or instances only accessible “for those in the know”, just not the base system.
I’d file this one as a bug, or at least a feature request.
This could be something about it, give it a 👍 if you have Github
Having remote links load on my local instance so I could interact with them would be awesome. Even better if my instance would fetch a remove posts & comments so it would really look like one unified platform without missing remote information.
I think you can hold alt to select link text without accidentally clicking the link, or at least part of it.
Today I learned
deleted by creator
I’ve been going to homeinstance.tld/c/[email protected] and subscribing that way.
I swear there has to be a better way though…
As I understand it, there is a way to format links that will open the community on your current instance. I’ve seen people say that, anyway, but don’t know what that link format is.
I can’t even find the community I’m looking for from lemmy.world . I search for [email protected] and it says no results
If you’re the first person to subscribe to a community from your server, what you need to do is go to the community search, switch from “Communities” to “All”, then paste in the full URL (https://lemmy.ml/c/worldbuilding)
I know it’s not great, but keep in mind that Lemmy just increased their userbase by 12-fold overnight and it’s a 2-man dev team. This isn’t some glossy corporate product, and there will be teething issues.
I always enjoy watching small teams completely run circles around giant companies. Y’all are doing great, we all thank you.