this post was submitted on 03 Jul 2023
2 points (100.0% liked)

LemmyFund

38 readers
1 users here now

Discussion about the LemmyFund Proof-of-Concept application.

Links

Rules

  1. Please be civil
  2. There are no dumb questions

founded 1 year ago
MODERATORS
SirQuackTheDuck
[email protected]
2
Welcome to LemmyFund (feddit.nl)
submitted 1 year ago by [email protected] to c/lemmyfund
4 comments fedilink hide all child comments
 

Hey all,

Welcome to the LemmyFund community. This community is made to discuss how we could shape a good, privacy-friendly method to fund your favourite instances and communities.

The code will be available on GitHub, but the main discussion will be held in this community.

Please have a look at the docs repository to get an idea of the main hurdles I'm trying to tackle. Any help is more than welcome!

Thanks all,

SirQuack

top 4 comments
sorted by: hot top controversial new old
[–] [email protected] 2 points 1 year ago (1 children)

Hello :)

To start, I'd like to clarify if there actually is a problem which is in need of a solution.

From what I've seen so far I've got the overwhelming impression the current donation systems work. I also think we have more than enough instances / capacity in the network.

I'm a bit worried about introducing a payment system which could be gamed for profit, and I'm worried about introducing a capitalist mindset into an ecosystem, which (as far as I can tell) works fine without.

But all that hinges on the question if there is a problem which needs a solution. I'm not claiming that's not the case, I just don't see it yet.

[–] [email protected] 2 points 1 year ago (1 children)

Hey,

Good questions! I don't see this as a replacement for donating to your instance directly, and those systems do work as they should (minimal fees, maximum impact).

The problem

The problem I'm trying to solve is that authoring servers/networks may not get funding if everyone is supporting their "home instance", but consuming from other instances.

For example, lemmy.world is very popular, and has a bunch of donors (myself included), but other instances host (expensive) content too, which might not attract as many donors (NSFW servers might fall in that category).

It might be that I'm completely wrong and data is duplicated all over the place, that should keep the bandwidth usage to instance users and server-to-server talk, which makes the cost-base more local.

That would still not benefit communities, which is where I want to be able to possibly head towards too, that you can fund communities you interact with directly, whether local or federated.

The gaming risk

This is definitely a risk I've considered and will always need to be guarded against. Some counter-measures I have in mind are:

  • Users funds will be split based on their own interactions, not because other users are big users of a big instance.
    • This should stop gaming a bit, since "spam users" won't affect how other users' funds are split
  • Users will recieve a detailed message per donation about how their funds will be split
  • Users can select how to split their money (by subscriptions or by interaction)
  • The company that should operate this will be registered under Dutch law as a non-profit foundation (either "ANBI" or "SBBI"), which means it cannot be for-profit and all books and records will be open to the public.

In the end, for me, it just tickles the "I'm doing my part"-bone to fund servers by usage, which is why splitting it up would then work.

[–] [email protected] 1 points 1 year ago (1 children)

Thanks for the answer, makes the idea more tangible for me.

It might be that I’m completely wrong and data is duplicated all over the place, that should keep the bandwidth usage to instance users and server-to-server talk, which makes the cost-base more local.

I believe that is the case, yes. Users interact with local copies of remote content, and federated servers share these interactions with each other. This also explains how defederated communities do not disappear, but you can only see interactions from other local users.

It might still be true that some instances have a higher operating cost than others, possibly even while having less users. While a video instance shares some load with others, it still has to host all of it itself. But don't take my word on this one, I'm still learning.

That point needs clarification from my point of view: What funding challenges do instances face, and are the challenges different for different types of instances?

That would still not benefit communities, which is where I want to be able to possibly head towards too, that you can fund communities you interact with directly, whether local or federated.

What funding do communities need? Unless we're talking about paid moderators, I think the instance pays all the bills.

[–] [email protected] 2 points 1 year ago

While a video instance shares some load with others, it still has to host all of it itself.

I think services like gifycat / imgur will pop up eventually to flatten out the data storage problem. Storing copies everywhere doesn't seem too sustainable (both economically and ecologically). I guess time will tell.

What funding do communities need? Unless we're talking about paid moderators, I think the instance pays all the bills.

I don't know, it's a long term goal. Moderators have been volunteers for forever with reddit and that worked out, but having some fund to give awards or compensations seems to be a good thank you. That's all not part of a first release.

My first version will likely be a "how much you want to donate" and a system that considers that input a full payment, giving the user an "invoice" in the end just to show where their theoretical donation would've gone.

From there we'll let the public debate decide if this is usable, sustainable and reasonable.