It won't stop the people interested in doing mass-manipulation of votes. There is no virtual difference between a bot that votes through Lemmy and a bot that votes through "pure" ActivityPub.
rglullis
No crazy scheme like this will work, because creating any form of anonymous account is super easy.
What we need is to add a layer of trust assignment and authorization system based on reputation, like what Fediseer does but built-in into the application. Then we can quarantine new users, contain their actions until the admin gives explicit approval, auto reject follows if they don't have at least N trusted people vouching for them, etc...
Creating a fake instance to send out votes is cheaper and easier that you think. You need only one domain and a script.
It's public information available for instance admins and any sufficiently motivated person who knows a bit about the ActivityPub protocol.
Another thing to consider is that most servers only accept HTTP requests that are properly signed.
Maybe you can take a look at Fedify if you know your way around javascript development? If python is your thing, maybe my own Django ActivityPub code?
I have an opinion about practically everything
Where is the /s?
they should perhaps create more content that I like?
Your account is less than 15 days old. You have 57 comments and one single post about politics. Not exactly the type of discussion that is interesting or valuable to the network.
Perhaps you should lead by example and post more of the stuff that you like? It's very easy to criticize, not so easy to actually show up and do the stuff that is needed.
You will need a server that can serve you actor URL and public key. Even a static webpage will do. If you have that, then a simple curl could be sufficient.
Not the case. I have this laptop for 2 years or more...
I think it will be okay if you and I stick to our remote accounts for this type of community, but I can create you an account there to act as mod if you prefer.
In any case, it would be nice to have this type of setup (Lemmy instance hosting community and mods on different mbin/lemmy instances) to see what are the real challenges for moderation at scale. I still have on my backlog this idea of working on an unified moderation tool.
Flip the thing around. Why should we introduce some dependency on a technology when the "simple" system is already very effective?