while i don’t have any specific opinions about this that other people haven’t addressed, i just want to flag up something;

How this could be enforced? No voting from the All and/or Local feed. Seems easy and straight forward.

this seems unenforcable. as in, you can’t really tell where someone discovered a post from. yeah you can just remove the buttons from those views clientside and it’ll probably work for the majority of cases, but alternate clients or modifications to lemmy-ui can simply put the buttons back in (or in cases of unmaintained or differently opinionated clients, just not remove the buttons at all). the backend can’t really differentiate which view a vote comes from. federation especially can’t differentiate which view a vote comes from.

What I’m more worried about are posts relating to news mainly. Where even if the immediate first level comments are fine, there are threads that get out of hand really quickly.

I agree that while posts inherently designed to be controversial may not benefit from Active considering the influences voting has (though me being on an instance that has downvotes disabled may be influencing my view here!), Active may make it significantly easier for an otherwise innocent post to devolve into a flame war.

The main excuse for this kind of algorithm seems to be around “promoting discussion”, but in my experience tech that’s intended to promote discussion does inherently promote flame wars too, as they’re extremely difficult if not impossible to distinguish without a human in the loop. I’ve attempted to write something about this on the microblogging side of the fedi, directly influenced by this post

I’m just throwing this out there but having the default sort incentivize comments seems like it’d highlight posts meant to cause flame wars… Is that what we want out of this platform?

well I just checked and while “sync contacts” did not turn itself on, “allow contacts to add me” did. there’s definitely something going on

I’m just saying the “steamlib proxy” concept you mentioned is not only possible but has at least one implementation. The fact that Wine isn’t true isolation doesn’t really change anything (maybe other than the exact details of the underlying IPC channel, which could use some kind of optimized shared memory magic) unless I’m misunderstanding something.

Some games will probably actually rely on Steam, like for achievements or something. For those…If there are a substantial number of Mac games that won’t work in a 64-bit environment, I am wondering if it is possible to make a “steamlib proxy” – basically, have a 32-Mac VM, run the game in a VM, but have Steam running in a 64-bit host environment, and just relay calls to a process launched under the host environment that uses the host steamlib to talk to Steam. Valve presumably isn’t gonna set that up as a supported environment, but I wonder if that might be a viable open-source project.

I think Proton has something of that nature, so games running inside Wine talk to the native Linux Steam binary.

DNS blocking is the most unreliable way of blocking youtube ads you can imagine.
you could write a script to OCR your entire screen and click skip ad and it’d be more reliable than DNS blocking

One of the reasons I use containers instead of installing things directly is that i can completely uninstall a service by deleting a single directory (that contains a compose.yml and any necessary volumes) and running a docker/podman system prune -a

or that i can back up everything by backing up a single “containers” dir, which i could have on a subvolume and snapshot if i wanted to

systemd/quadlet on the other hand makes me throw files in /etc (which is where you’re supposed to put them, but ends up resulting in them being tangled together with base system configuration often partially managed by the package manager)

The Solution™ to this is configuration management like ansible or whatnot, which needlessly overcomplicates things for the use cases i need (though they’re still useful for getting a base system “container ready” wrt ssh hardening and such)

tldr: i want my base system to be separated from my services, and systemd integration is the exact wrong tool for this job

In Logseq, everything is a nested list. This feels like a limitation, but I’ve been preferring it. The decision is made for you: you’re going to jot this information down as a list. So then you just start writing it.

Oh - this sounds interesting.

Whenever I needed to jot down any notes I’ve been finding myself just writing plain .txt files with bullet points, and trying tools like Obsidian or TiddlyWiki I always ended up being overwhelmed with the amount of stuff I could do (and with all the customization options) that I never got around to actually writing things down. I’m definitely gonna look into how Logseq works.

(Although I have to say, their website does look a bit “too hype-y” for my liking. IDK how to explain it, just a gut feeling. Still, at least it’s FOSS so it can’t be too bad)

Oh no it’ll federate alright.

The thing about ATProto is that unlike AP they don’t seem to expect each instance to have it’s own community with it’s own rules and vibes. They seem to be using federation just as a way to “scale up”.

If they can get any non-bluesky-the-company folk to create instances then that’s just scaling they don’t have to pay for and a convenient legal scapegoat for the inevitable consequences of their lax moderation. Why wouldn’t they federate?

If you have Authorized Fetch turned on (or if you use something like GoToSocial which forces it on) this won’t work (as far as I’m aware anyway) as neither Lemmy nor Kbin support it completely.

To be fair join-lemmy also advertises the software as “blazing fast” with “powerful mod tools” and if there are two phrases I’d associate with Lemmy those aren’t it.

(OK compared to Mastodon Lemmy is blazing fast, but that’s cheating considering of how slow Mastodon is)

it can be pretty useful for really specific cases, but i’m not exactly sure if this is one of them

the allowed instances list acts as an allowlist, meaning you’d be defederating yourself from the rest of the fediverse (and only federating with the instances you allow).

if that’s what you were going for ofc it’s your right and i sure won’t stop you from doing that, but i feel like you’ve misunderstood what it is.

The Synapse and Postgres containers are using about 2% CPU and 1.2Gb of RAM combined. If your strapped for resources, check them out.

with 1.16G i can run dendrite and iceshrimp (+redis and +sonic, shared caddy and postgres between both). i wouldn’t call synapse the “strapped for resources” server.

i’d say 6-7k users would be a better figure, but i do agree with your overall point. anything above 10-15k just hurts the health of the federation in the long term and ends up creating “untouchable” servers. even worse on lemmy where communities are also bound to instances.

though now that i’m thinking about it, a better metric would be active users rather than total users

if you have a hard time choosing between Gitea and Forgejo I recommend picking Gitea for now, as they haven’t done anything bad just yet, but if they do Forgejo supports migration from Gitea.

iirc there isn’t an official way of migrating the other way so if Forgejo fucks up you may end up out of luck

it’s partly because everything has public/private certificates, but also partly because there isn’t much synchronization going on after the initial “push”. if you shut an instance down and modify the database directly without informing other instances (say, you remove an account) then other instances will not be able to tell and will drift out of date, essentially making that specific thing unusable for any instance that has previously interacted with it. if you expand that out to, say, wiping and re-creating an entire database, then you end up with so much uncertainty that you may as well start over from a fresh domain

You’d have to be significantly more careful with backups, as it’s really easy to effectively “burn” a domain from federating over AP ever again (at least to instances that federated with it before), but otherwise it should be reasonably automatic as federation gets implemented piece by piece.

I’m not sure on if we’re talking about the same thing here but as someone who disables JS by default, any federated service without some form of SSR is inherently clunky to use. Half my allowlist is Masto instances I’ve barely visited once or twice because website boy is a jerk and ruined a reasonably well functioning SSR UI.

I don’t get why the language would make a difference in how you deploy the frontend. You can already host the BE and FE on different hosts with the right reverse proxy config. This just replaces the Node in lemmy-ui with Rust+Leptos.