I don’t think its too bad, but it probably depends a lot on a lot of factors.
Since I first started my hardware got a lot stronger, and nextcloud, php, and mariadb have all improved and so my experience has gotten pretty decent.
Remember though, there’s a ton of biases here, so I could be wrong…
Nextcloud Notes has become my go-to (Oh look, SJ is advocating for Nextcloud again! How original!)
I’m cheap.
So far, Conduit is the only answer for me, since I don’t own any quantum supercomputers.
One big difference between the json requests and a user callling for the site directly is your instance pulls all the data all the time, whereas a user only pulls the data they use themselves.
I think it depends a lot on the federated service.
For mastodon, you follow individual users, so if there’s a million users or ten million or a hundred million, their instances will only be contacting other intances they’re federating with so it’s quite scalable.
For Lemmy, you follow communities, so every server pulls all the posts and comments the common community. This means that for an instance like lemmy.world hosting lots of different big communities, every new server hammers the one central instance.
A strategy for improving the situation I think would be to spread the load. Instead of everyone piling into megacommunities, if people spread out into smaller more tight knit communities over many different instances. Of course, this isn’t really compatible with the purpose of having communities like that.
It does seem to suggest that ActivityPub isn’t necessarily the most appropriate protocol for this purpose, even though it’s what was used because it’s the de facto standard on the fediverse.
Depends on the specific instance. Some services run where they are natively accessible through tor, but most don’t.
Just remember that ActivityPub is a sharing protocol, and individual admins are fully capable of seeing everything. There is no end to end encryption, everything is stored in plain text.
You don’t need a local DNS server to set up https, but you do need a domain name. If it’s something that you wanted to pick up, you can buy them at a number of different places and you’d have to set up a mechanism to make sure the IP address referenced is the correct one. You can either do that by having a static IP address or by setting up some form of dynamic DNS. Then you can use letsencrypt to set up https.
Okay so here’s I think the core of your question though: the only way that someone outside of your network can access your nextcloud is if you have set up the server to be accessible from the outside world. You would have to go into your router and forward Port 80 to the local IP address of your nextcloud server. If you don’t do that, then it will only be accessible to the people inside of your network. Rotors do something called Network address translation which lets many devices on your local network connect to the internet despite only having one external IP address. If you’re accessing the server using a 192.168 address or a 10.x.x.x address you are already using the internal IP address and not your external Internet IP address so you’re likely safe.
One neat trick because remembering IP addresses is a pain in the butt is the hosts file. On windows it’s in c:\windows\system32\drivers\etc\hosts and you can set a hostname to immediately resolve to a certain IP address. It’s particularly nice because it’s free, it’s fast, and once you set it you can forget it.
My websites are on the public internet, but I use the host to file to point them at the internal IP address because that way I can directly connect to my servers even when the internet is down.
Apparently you’re not aware that science is in a deep crisis right now on several fronts in part because academic researchers know full well they have to give the “right” results in order to advance their careers.
Because the internet is not your friend and something exactly like this will happen. Then someone will take something you said totally out of context to try to get you canceled.
No, I’m saying it does work, but other clients might not.
Besides lacking spaces and some rooms not letting you join, (and the lack of admin tools) the only big issue I find is that you plan to run something other than Element as the interface, you’ll have to test it because many matrix clients expect synapse or dendrite and won’t start with anything else. I’ve run fluffychat, I think kchat(whatever the kde matrix client is), and nheko, they all worked well with conduit.
My experience has been that dendrite and synapse totally maxxed out the server I ran it on (100% cpu utilization for days on end), so I run conduit.
The one downside of conduit is it’s a bit behind, so it doesn’t support all the latest rooms, and it doesn’t support spaces yet, and it has minimal admin tools so you’ll want to create all the accounts you need then close logins because bad actors will try to create logins and get you banned from half of Matrix. That said, I can tell you that even on my piddly little server (an Intel Atom D2550), it runs Conduit, ejabberd, nostr, and lotide, and the server basically sits idle. I can’t speak of bridges, unfortunately, because I don’t really use them.
This is the guide I used, it worked well to set things up:
Yes. This isn’t big tech where tech daddy is sitting over your shoulder reading your messages to sell your info to advertisers.
The article, as they say, “Assumes facts not in evidence”
Who puts their real name into their mastodon account…?
It’s also sort of an egocentric action – “I don’t like some users over there, so nobody should be allowed to interact with any of the users over there”
The threadiverse will continue to grow, and there’s going to be people of all kinds joining up. We need to figure out how to coexist as individuals because you can’t just have every instance defederate with everyone else because there might be a bad user.
So there’s 2 things, I think.
Does your bios allow you to boot from SD card? If so, then you can boot from the SD card and so you can install software onto the SD card directly.
If you can’t boot off of the SD card, then perhaps you can install all the software on the SD card and then install a boot manager on the main drive. In this way, you boot off the main drive, then let the boot manager deal with loading the software.
You might be disappointed by the performance of software running off an SD card, mind you.
I think at first it didn’t. As much as history tells us everyone was watching, the reality is that he had some long hard runs through many areas where he wasn’t getting much recognition and wasn’t getting many donations either. Someone from British Columbia walking through Quebec faced some serious challenges in 1980. The big cities speak English, but a lot of the smaller communities along the highway are very French and don’t really speak English.
That thankless run across Quebec, where he was getting run off the road constantly by drivers, barely any attention, and few donations, it must’ve been a herculean task.
If I had to guess, I think it’s probably the Cancer Society that was sponsoring him was able to pull some strings and make his arrival in Ontario on Canada Day (July 1st) a really special event, he met with the prime minister of Canada, prominent sports teams, and virtually anyone he thought would help the cause of cancer research. He became something of a celebrity after that.
I’ve found the biggest thing isn’t any real resource. My instance runs on a core 2 duo with 4GB of RAM, and I really try to get it to waste memory and barely fill the 4GB.
The thing is your instance will be blasted by all the other instances you subscribe to. If you subscribe to too many big communities you might find you’re locked out during peak times, but it should be just fine as long as you’re not crazy with follows like I am lol