gyptazy.com is a Fediverse instance that uses the ActivityPub protocol. In other words, users at this host can communicate with people that use software like Mastodon, Pleroma, Friendica, etc. all around the world.
This server runs the snac software and there is no automatic sign-up process.
Wondering how and why? With @BoxyBSD@bsd.cafe and @BoxedTux@mastodon.social, I’m already running platforms that provide free VPS instances for educational purposes, learning and testing for years now. I guess, providing real AS and peering services (maybe even sponsoring LIR in cooperation [but this would cost money]) could also be very interesting for several ones, apart from the DN42 network. Happily looking for feedback…
#isp #network #networking #ipv6 #as #autonomoussystem #ipv4 #free #opensource #devops
I'm learning a bit about #networking right now.
Can someone explain to me why exactly the #Matrix folks chose to build their client/server and server/server protocols on top of #HTTP instead of writing their own protocol on top of a lower layer network protocol like TCP/IP or UDP?
The FAQ mentions that they chose HTTP because it's a common, widespread and well understood baseline. But since you still need to write software specifically adapted to the Matrix protocol anyway, isn't it kind of irrelevant that a lot of other applications use HTTP? It doesn't really explain why they even needed to 'ride' an existing application layer protocol rather than developing their own.
After all, whoever wants to develop a Matrix client or Matrix server software needs to specifically implement the Matrix protocols anyway, so it's not like you can just use an existing "HTTP client" or whatever to use Matrix. Would it really make it so much more difficult to develop client or server software if they had to implement an application layer protocol rather than piggybacking on HTTP?
I'm wondering mostly because HTTP, being a hypertext-focused protocol designed originally to exchange static hypertext documents, doesn't seem like the ideal first choice for a real time communication system, other than that it is popular I guess. It seems like it'd require quite a lot of wrangling HTTP into a shape it was never meant to adapt rather than writing your own protocol. It kinda sounds like trying to create a video file format on top of JSON because a lot of people know how to use JSON.
If I designed my own messaging protocol, my first intuitive choice would be to write specifications on top of, say, TCP/IP that my client and server software would understand. Would that hypothetical course of action be a mistake?
RouterOS 7.23.2 release notes:
!) fixed a service security issue, home user with default config not affected, but we recommend the upgrade for all users regardless;
so... we fixed a security vulnerability, but we're not going to tell you what it is, so you have no idea how urgent this update is or how exposed you are or if you're even affected at all.
thanks, i guess?
(rumour has it this is a memory disclosure or potential RCE in pptp/l2tp/ppoe...)