For context, I am the top donor at my instance, I recognize that there is a need for funds. BUT, I believe it’s important for the fediverse to be accessible to everyone regardless if they have the funds for that.

Dopamine received, initiating hyperfocus protocol!

As a rule of thumb, we’ve observed that a team of 5 trained moderators appears to provide ample coverage and redundancy for servers of about 1,000 active users

That’s a fascinating bit of information. I would expect 5 moderators to provide coverage for more users. I am wondering how they came up with that statistic (will update the comment if I find an answer).

Remember that offliine/IRL community management experience can be just as important as online experience

Interesting idea, wondering what’s the IRL presence of the fediverse…

If you’re building toward participatory or democratic governance, consider establishing a proposal and voting system (some teams we spoke with use Loomio, but multiple options exist) for major policy decisions.

That’s soooo important, I love when communities create polls to decide on policy changes.

Avoid promoting brand-new members unless you already have a pre-existing relationship with them

I have followed some discussion on multi-level hierarchies on the fediverse, wondering if there are any instance implementing that…

Consider charging for accounts or offering paid memberships.

Hell no!

We hope there will be more resources available in the future, particularly tooling around legal compliance. This is one of the big infrastructural gaps we point out in our main report

That’s a big issue, I would be interested in hosting an instance available to other people, but I don’t want to end up in jail and I lack the resources to make sure that won’t happen…

That was an interesting read, it seems there is an in-depth analysis of the report here.

Can someone ELI5 so I can get enough dopamine to go read the whole thing?

Not sure if you are being sarcastic or not but I found this review.

https://www.mdpi.com/2073-8994/14/5/858

I had done some research about a year ago, but I don’t have the papers saved.

While the consumption for AI train can be large, there are arguments to be made for its net effect in the long run.

The article’s last section gives a few examples that are interesting to me from an environmental perspective. Using smaller problem-specific models can have a large effect in reducing AI emissions, since their relation to model size is not linear. AI assistance can indeed increase worker productivity, which does not necessarily decrease emissions but we have to keep in mind that our bodies are pretty inefficient meat bags. Last but not least, AI literacy can lead to better legislation and regulation.

If you are worried about duplicates, aka a single bot spamming multiple votes, then that’s feasible to mitigate.

If you are worried about multiple bots spamming one vote each, that’s harder to mitigate and it comes down to how the instances handles bot accounts in general. IMO it’s best to ignore the bot problem and instead focus on designing a vote weighting system that favors similar instances.

For anyone interested, there are a few papers on cryptographically secure voting, where both voter anonymity and election integrity are preserved.

Most designs consider three separate entities, where if you accumulate the information between those entities you would be able to identify a voter and his vote, but each entity on itself does not hold enough information.

Follow people from other instances, if you find them interesting, chances are they will find you interesting and follow back.

Use tags, especially popular ones.

Gitea and forejo are doing some amazing work

Probably a lot less, keep in mind that whenever it answers a question the whole model is traversed multiple times, going through multiple GBs is not possible in the matter of seconds the model answers.

They made meow meow beans?