they likely blanket banned/filtered the .zip tld due to the rather large security problems it brings. I had the same problem with my instance initially before I explicitly whitelisted lemmy.zip

I’d be very careful with those words given that lemmy.world is rather trigger happy with defederations.

It is almost impossible to completely censor away content and anyone can bridge blocks by creating a new intermediary instance where they can view content of mutually blocked instances but it’s not like you see all the fediverse content on any instance.

Matter of fact you just need to look at the blocklists of most instances to get a pretty good guess at the other, less legal, Fediverse out there running almost in parallel to the popular one. But since almost every instance has those blocked you don’t see any of that shit in your feeds. Highly censorship resistant, but ultimately an individual is still at the mercy of their instance admins.

> make a new messenger using a niche protocol > new users choose your messenger because it is objectively the best after you dumped unreasonable amounts of cash into it > userbase grows, in large parts because the small messenger is interoperable so you can say “hey, if other company wanted to they could just implement [protocol] for you, we are already doing that” > once userbase reaches critical mass, pull the plug on the protocol > users with long chat histories and contact books are now more or less stuck on your platform whether they like it or not because getting people to switch suddenly means two messengers instead of one for them, not a good proposition to make.

XMPP did take off while it was in Messenger, Facebook decided to kill it with its superior reach because it was a step-ladder rather than something actually useful to them. Facebook will absolutely use the Mastodon interoperability as a marketing trick “Hey guys, if you have friends that don’t like threads they can use another platform and still talk with you”. They’ll use it to distinguish threads from twitter until they feel like they don’t need it anymore. Then they’ll find some sort of technical excuse and pull the plug on ActivityPub support.

If they are having instance owners sign NDA’s they must have pre-selected who to talk to or only target Mastodon because feddit.de just pre-emptively defederated threads.net after users made the owner aware of the news and it’s one of the largest instances (even more so if you go by MAU which is arguably what Meta would be intersted in).

The way I see it have them show through actions that they won’t EEE for a good few years with whoever is willing to risk it. If they don’t show any signs of EEE then, I might reconsider my stance on federation with them. Until then I’ll keep threads.net on the blocked list of my instance.

yeah you’re right, wasn’t fully awake I guess

We already sort of did, it’s called tauri

Just piling onto this I really hate how it’s always portrayed as if ftl would somehow break causality. I cannot for the life of me figure out any way in which something travelling faster than light would do that. Travelling faster than sound doesn’t break the air eithe (well it sort of does but the air is still there working as air once the sonic boom has passed it)r.

Bit of a tangent but I get the feeling a lot of scientists are stuck revering the old geniuses a bit too much. Einsteins formula is basically taken as gospel, to suggest it might be inaccurate is seemingly treated as heresy and I don’t think that’s a good thing.

Newton’s theory of gravity was also revered as undenoable fact but lo and behold it was severely inaccurate. What makes these people believe it’s any different with Einstein and co? (arguably this could be down to the media distorting sentiment among scientists but that only improves things marginally)

Not sure if possible but have you checked whether the running user is an application property? If so you could maybe manage it via application or window rules.

Can’t check rn because I’m away from my PC but it’s worth a shot checking imo.

cosmically unfortunate timing, package maintainer probably just took a nap between updating most packages and fixing the leftovers and the resulting gap caused this mayhem

I would definitely recommend against -Syudd with kde-unstable, the packages can change drastically (as was the case here) breaking the entire system if dependencies are ignored (as happened here)

Update: the bad packages got updated, you should be fine just installing them from kde-unstable again

Hopefully helpful:

you need the following packages for kde-unstable:

• plasma-activities
• plasma-activities-stats
• libplasma

Your Update likely removed these from your system breaking plasma6

The following packages are likely to blame (this list is likely incomplete):

• kweather (I removed it since I don’t use it)
• dolphin (as suggested above switching to dolphin-git should help, alternatively installing extra/dolphin will install the plasma 5 version bypassing the dependency problems)
• kio-extras (same fix as for dolphin)
• sweeper (see above)

Edit: if you have terminal access you can try the following (no guarantee it’ll work, I think it should but no guarantees):

• pacman -Rdd kactivites kactivites-stats plasma-framework
• pacman -Syudd libplasma plasma-activities plasma-activities-stats

What I did (switch to plasma 5 versions):

• pacman -Syudd extra/dolphin extra/sweeper extra/kio-extras extra/kweather

What you could do (switch to git):

• $AUR_HELPER -Syu dolphin-git sweeper-git kio-extras-git kweather-git

Codeberg is iirc the main entity behind it, at the very least they are using some of their funds to support it

see my reply here

I’ll try to summarize:

• Gitea is managed by a For-Profit that apparently popped out of nowhere -> profit motive conflicts directly with FOSS and since the corp isn’t well known it must be assumed acquisition was solely to make money
• Gitea now requires Copyright attribution, meaning if you push code to Gitea in an existing file it ain’t your code anymore -> omega level oof for a FOSS project because this essentially kills any upstream contributing (as seen by Forgejo deciding to stop their contributions)
• This Cloud Service being offered when Self-Hosting Gitea is really easy, again -> profit motive conflicts with FOSS but now on steroids because a “core” feature of their service will limit their ability to make more money

I have honestly no idea what the GitLab devs did but their service is such an incredible memory hog it’s insane. Obviously GitLab has a pages service tacked onto it but my GitLab instance (mostly legacy but a friend still uses it so it keeps chugging along) eats a whole 5GB of RAM while my Forgejo Instance only uses 200MB. I have no idea where all of that memory is going because it sure as hell isn’t going into responsitivity. I’ve no idea if I configured something wrong or if it’s GitLab pages but it’s still excessive

huh, would you look at that. Pretty stupid move and something that makes this entire thing even more suspect. Glad I picked Forgejo over Gitea

smaller company

true but then again that isn’t quite what I meant with my comment. For an individual looking for a self-hosted forge GitHub just isn’t really an option. Pricing aside having to go through a sales channel and then likely not having full control over the software stack is not what individuals look for when they want to host a private git service

Forgejo has different development priorities but feature wise they should be identical since the Forgejo devs also push their code upstream into Gitea