You're viewing a single thread.
I think there will need to be a re-invention/ re-write before the 'real' lemmy happens.
The issues around discovery, identity and communities being attached to a server, and how this impacts the math of the networking aspect of the whole concept are really what is preventing Lemmy from not only beating social media like Reddit, but representing a major improvement. The project has grown enough to make that kind of re-investment of time more palatable, but as lemmy currently stands, the protocol is in the way.
We don't need to beat reddit. We don't need to beat anyone. There are no investors or shareholders. There are no stock listings. There are no ads or addictive algos. We are fine as we are now. There is no need for exponential growth. Lemmy should simply be.
We do need to continue growing at a natural but sustained rate. 50-60k is not a healthy place to stop and there's still a lot of low hanging fruit development-wise.
Its not clear to me that the issues I'm identifying can be resolved with a new release. Id love to have the discussion, but it goes deeper than just discovery.
I think the best example of how to 'lemmy' properly, or in a way that doesn't create 'wasted votes' (in the gerrymandering sense) of content, is the startrek lemmy. The focus on a niche topic and own it entirely. Theres no point in having lemmy subcommunities based on startrek because the startrek lemmy is so great and makes such great content.
Contrast this with lemmy.ml, lemmy.world, most of the other big instances. Mostly redundant sublemmys. Because of this, the quantity of content is very thin, and because of this, good content is less likely to gather momentum. A way to group sublemmys across instances could work, almost like a kind of sub-federation (like, on lemmy.world, maybe lemmy.ml/c/memes is federated into lemmy.world/c/memes. any content posted to one is 'considered' posted to the other). It seems relatively straightforward (not a trivial lift, but reasonably possible) to merge the two RSS feeds.
The same issue exists with discovery. Things are too diffuse in lemmys current network structure.
I've gone through some graduate level coursework on network mathematics, and I work with networks in a very different context, but I don't exactly have the math skill to write out a proof on this. This paper outlines the basic concept, but to extend it to social media, we basically get much much better content in the form of submissions, comments and discussions, with super-connectivity. Its kind of fundamental to social media graphs and there are some clear barriers in lemmys design that prevent it.
A way to group sublemmys across instances could work
I think user side multi-communities whose definitions are easily shareable so that you can basically set up a bunch based on a curator’s or moderator’s recommendations … would do the trick nicely and be pretty achievable technically. Frankly, I feel they’ve been a long time coming and hope that they’re in the works already.
I'm not sure its in the works, but it needs to be very well thought through prior to implementation.
Like, what makes sense to me would be a way to 'sub-federate' two or more RSS feeds into one. IE, if lemmy.ml and lemmy.world are federated, lemmy.world has the option to 'sub-federate' c/memes from .ml with c/memes from .world, or at least the moderators would have that option. You may only want there to be a 1:1 with sub-federation (IE, you don't want many .world /c's subfederated with one .ml /c or you would get a (possibly extreme) duplication of content. But this could be option and up the the lemmy instance to decide how to configure.
Regardless, the 'wideness' of the way the network is set up with many nodes and relatively few edges is the primary issue. It can't be resolved with just 'more engagement'. Reddit doesn't suffer from this issue as much because there its only one order of dilution (one reddit, many subreddits). Lemmy suffers more because there are two orders of dilution (many lemmy, many sublemmys). Its also important to note here that Reddit did not have subreddits to begin with. They came later, which helped them build the critical mass to overcome the diffusion problem.
I’m not sure its in the works
Possibly ... but it's been spoken about by the devs before, so it's on their radar at least. It may have even come up in their AMA? I know a way of sorting that surfaces smaller communities definitely did come up and is definitely in the works.
Otherwise, I'm not sure you're convincing me.
It seems to me that you have to start by establishing that "parallel" communities (ie different communities with similar or identical focus) here are worse than on reddit. Unless there's a large amount of complex defederation going on, I don't see how the decentralisation substantially worsens the effect the existence of parallel communities has on "engagement" compared to the situation over on Reddit. And, as far as I understand, such a large and complex defederation/federation network has not happened. The most significant example of defederation is probably the
beehaw-lemmy.worlddefederation, where beehaw only has about 3% of lemmy's active users.So I'm not sure your
two orders of dilution (many lemmy, many sublemmys)argument follows. THe196community over onblahajseems like a good example, where it has 5 times as many active users as there local to the whole blahaj instance. thestartrek.websiteis in a similar boat. Federation seems to be working fine. And while there are duplicate communities lying around, I don't think that means that people aren't naturally flocking to where they feel is the best place to be for things while allowing diversity to exist.So, if parallel communities are fine, or even good (which is my take, at least to an extent), I'm not sure there's a sufficient argument for the need to implement sub-federation on the backend. I'd bet it would be tricky and add a whole new kind of entity to sort out.
Comparatively, the multi-communities idea, I'd guess, is a much more natural extension as it's really just allowing a user to have multiple sets of subscriptions, which is an already established process/feature.
Also, I'd wager that putting flexibility in the hands of users rather than community moderators is probably a better way to go.
Comparatively, the multi-communities idea, I’d guess, is a much more natural extension as it’s really just allowing a user to have multiple sets of subscriptions, which is an already established process/feature.
I guess that's where I see the issue and maybe were talking past each other a bit. I think if its community moderators that 'subscribe' their sublemmys to multi-communities, that could work. But if its on individual users, its not going to solve the fundamental underlying issue of discovery.
I hear you. I appreciate you breaking down the conversation too!!
I guess I’d hope discovery could occur through wide spread multi-community suggestions, ideally posted in community sidebars too ( though that’s probably unlikely). I’d also hope the cross post interface gets better, with all cross posts listed (not just those that share a URL), and the ability to view all comments from all cross posts together (maybe with some options around whether you subscribe to them or not).
Beyond that, I’m still not sure we’re in a worse position than Reddit was, apart from currently having a small size?
Oh, overall I think lemmy is way better than reddit, even now. But the bigger the fediverse gets the more consequences come from drastic changes. Good to have these discussions now going through the growing pains under a relatively small user base.
These things absolutely can be fixed in an update but I think we both know the devs have no intention of going in that direction.
Its not clear to me that they can. Some of these problems relate back to activity hub.
And I just want to be clear, I think what we have currently based on activity hub are a great, proof of concept 1.0 (mastadon, lemmy, so many of the others), and that these lessons learned will feed back into activity hub 2.0, at which point there can be a material 'taking back' of the internet from corporate interests.
But right now, its not clear to me the current protocols or systems are where they need to be.
A way to group sublemmys across instances could work
there's a github issue for this
https://github.com/LemmyNet/lemmy/issues/818
it'll probably happen in stages, with it first just being a list of communities, then maybe they can get a name and be shared with multiple users and you could follow them like a normal community
I don't see any reason it would require a rewrite, currently a community is an Actor that Boosts the posts directed to it, a multi-community could also be an Actor that boosts any posts going into the communities/Actors that it follows
I think the best example of how to ‘lemmy’ properly, or in a way that doesn’t create ‘wasted votes’ (in the gerrymandering sense) of content, is the startrek lemmy. The focus on a niche topic and own it entirely. Theres no point in having lemmy subcommunities based on startrek because the startrek lemmy is so great and makes such great content.
I found this paragraph pretty confusing, probably because of uncommon terminology.
With "the startrek lemmy" you refer to one specific instance? Which? 'Lemmy' is commonly used to refer to the platform, or the software.
"lemmy subcommunities" refers to communities? Like https://lemmy.world/c/fediverse? Later you use the word "sublemmys". Does that refer to the same thing, a community?
Overall the suggestions make sense for me. But it isn't as trivial to solve, because of politics and policies. Maybe the startrek instance has great content, but does not allow hate speech. So "free speech" ultras might see demand for a startrek community on a "free speech" instance. Or hate speech is allowed, in which case the same scenario happens for everyone else.
Another line of division is the bot question. Are bots allowed to make new posts? Are bots allowed to make new comments?
What's the moderation style?
People are diverse. A one-size-fits-all-solution will likely leave some demands unsatisfied. If that portion is big enough, it justifies redundant communities. And there are many more reasons to possibly see redundancy as a good thing.
People who like a centralized approach can flock to the biggest instance or community, and others can do their thing. Both can coexist. What would be nice to have is view-grouping of communities, from the reader's perspective.
startrek.website was started during the exodus from former members of the startrek subreddits. All the communities (afaik) there are focused on startrek in some way. They post some great content.
The concept of 'wasted votes' in gerrymandering is when you can pack voters into specific districts based on voting demographics and population, to make it such that you can manipulate the outcomes of elections, or make districts that wouldn't be competitive based on the population, competitive for one specific party. Its a kind of efficiency measure of an election. It has a corollary in these threaded style communities, where if engagement isn't rewarded (be it upvoted, commented on, or submitted), the reward cycle doesn't happen and no additional content or engagement is created. In these communities the wasted vote concept applies because now engagement is spread out over redundant posts, comments, and lemmys. If you have two threads representing effectively the same thing at the same time, that could be considered a wasted vote, because now engagement is split.
And yes, I've been using sublemmys and communities interchangeably. I don't know that a common parlance has evolved yet. There have been a couple threads discussing this.
Overall the suggestions make sense for me. But it isn’t as trivial to solve, because of politics and policies. Maybe the startrek instance has great content, but does not allow hate speech. So “free speech” ultras might see demand for a startrek community on a “free speech” instance. Or hate speech is allowed, in which case the same scenario happens for everyone else.
I think this is a great point and I see why its warranted. There is another argument further down that's similar to yours.
Yeah actually, being better matters. Projects like this die with stagnation and attitudes like yours. Sure they'll limp around for years with some core die hards or niche communities, but thats (see: fark, SA, usenet, CL ads, etc..), but that's not the point. Lemmys design is working against its self. Its not clear that it can be fixed under 'lemmy' as it currently exists.
The experience can be much better and there is a clear path towards it. The basic math of how networks operate that creates this issue and its baked into the underlying structure of how lemmy was planned. If these improvements arent implemented, the platform will stagnate: all platforms that don't improve do this. They may persist but they fail to grow, and attrition is constant.
What's wrong with Usenet? Sure there's not a ton of users, but that's a good thing.
While I agree that we don't need to "beat" anything or strive for growth, I do think those things will happen naturally if the system is an improvement. And while lemmy's potential is great, there are challenges that come with federation, like those mentioned above. And those problems should be solved in time. Not to generate growth but to improve the system. Growth may follow