Searching yarn

Twts matching #http
Sort by: Newest, Oldest, Most Relevant

Ubuntu Desktop 23.10 ISOs Recalled Due To Malicious User Translations
Hours after the release of Ubuntu 23.10, Canonical has pulled the ISOs and is re-spinning them after user-submitted translations for the Ubuntu installer turned out to contain hate speech… ⌘ Read more

⤋ Read More
In-reply-to » fractalnetworksco/selfhosted-gateway: Route HTTPS traffic to local Docker containers through a cloud VPS over WireGuard. Ideal for self-hosting behind CGNAT.

@prologic@twtxt.net I do similar. Though probably much more simple.. I have CGNAT and use wireguard to VMs to punch through for stuff like HTTP/SSH from external.

And for SMTP I have smart hosts on the VMs that will store anf forward to my mailbox if the connection goes down.

⤋ Read More
In-reply-to » Hmm when I said "Wireguard is kind of cool" in this twt now I'm not so sure 😢 I can't get "stable tunnels" to freak'n stay up, survive reboots, survive random disconnections, etc. This is nuts 🤦‍♂️

@prologic@twtxt.net I find the L2 mode where you have one interface and multiple hosts to be tricky. Its best if you are trying to make a full mesh style. But then all hosts need to be able to see one another.

I have had more success using point-to-point connections where there are only two ends to each interface. It means you have a ton of interfaces and udp ports. but you can share the host IP across the interfaces. Add to that a simple router proto ala OSPF or RIP and you can navigate around not having a full meshnet.

I have dozens of localnet wireguard connections and many more connections to others that use bgp for route propagation.

⤋ Read More

Hmm when I said “Wireguard is kind of cool” in this twt now I’m not so sure 😢 I can’t get “stable tunnels” to freak’n stay up, survive reboots, survive random disconnections, etc. This is nuts 🤦‍♂️

⤋ Read More

More data contradicting the existence of “echo chambers”. As I’ve argued many times before, the concept of an echo chamber or information bubble is not real. The podcast below is an interview of an author of a study where they actually intervened and changed the information diet of 20,000 people (with consent!), then surveyed them after three months. They observed essentially no changes to the study subjects’ beliefs and attitudes. They also observed that the typical person, while they tend to gravitate towards people with similar political leanings, only get about 50% of their content from such like-minded people. They get the rest from neutral sources and maybe 20% from non-like-minded people.

Varied information diet + No change in attitudes when information diet is forced to be different = no echo chamber.

Listen to the podcast episode here

⤋ Read More