Investors: shut up and take my money!
Investors: shut up and take my money!
uBlock Origin is already less effective when running in Chrome than in Firefox. For example, it can’t detect CNAME cloaking on Chrome, while it can do that in Firefox. When Chrome finally enforce manifest V3, uBlock Origin will be even more neutered in chrome due to limited number of blocking rules.
I just checked my AMD box and tailscale there can consume ~15% of cpu time when the tunnel is under active use. When it’s not used it’s ~1.5%. But it’s a low power old AMD cpu though (AMD G-T56N), so I’m not use if it compares to Ryzen 5. On my intel machine, it’s ~5% when under active use, and idle at ~0.5%.
On my machine it’s consuming about 0.5% - 1.0% of cpu time, which is higher than zerotier in the same machine (almost zero).
Tailscale does a lot more things than just tunneling though. For example, on default installation it’ll catch all outbound dns request on the machine and route them through MagicDNS (100.100.100.100).
Whelp, nextcloud isn’t known for being fast. I don’t have hundreds of thousands of emails yet so I can’t comment on that, but one thing for sure is as you put more and more data on it, you’ll have to add more CPU and RAM to it or it’ll getting more and more sluggish.
I haven’t noticed any performance issue so far. I think they use wasm which help with speed. Too bad it’s not open source, but the fact it’s developed by a single guy working on it full time is actually very interesting, considering the webapp is actually work better than some apps developed by bigger teams. It can even edit PDF and gif!
I think using container instead of VM should be better for maximizing resource utilization in a raspberry pi. Instead of partitioning your tiny 8gb RAM into 3-4 VMs with even tinier RAM each, you can run a dozen of containers and probably still have some free RAM.
Believe it or not, NextCloud. It actually can work as an email client. And it can sync calendars, contacts and todo list too.
I always look for excuses to get more servers, so if you ask me, I’d say yes, get that new server. There’s no such thing as having too much servers since there are so many things I want to self-host.
I also regularly tear down my servers and see how fast I can set it up again. Keep my deployment scripts up to date.
photopea.com is actually pretty great, much easier to use than gimp with similar (or even better) feature set.
You can enable it from settings -> remote control in SmartTube app. Also, if you’re using YouTube revanced, the cast button might be hidden because it was not functional. You can re-enable the cast button on YouTube revanced from revanced settings.
When casting a video, you’ll have to open the SmartTube app in your Android TV first because it’s not automatically launched when you hit the cast button in YouTube revanced. If it’s in the background or not yet running, the video won’t play.
Usually there is a flag to disable muted video autoplay on every major browsers: https://www.technipages.com/turn-off-video-autoplay-firefox-chrome/
Safari has it too: https://support.apple.com/guide/safari/stop-autoplay-videos-ibrw29c6ecf8/mac
A while ago Libya suddenly requires all companies that use .ly domain to have a presence in Libya or have their domain reclaimed by the government. bit.ly (and other internet startups that use .ly domains back then) suddenly found themselves in a precarious position. It was pretty hilarious as .ly TLD was hip back then.
Consider how many system relies on being able to send you an email for verifying your login and performing password reset. Those who have control over your email address domain can trigger password reset for most of online services out there. Imagine if Google forgot to renew gmail.com and it falls to a wrong hands.
Here is the documentation, pretty bare though: https://docs.nextcloud.com/server/latest/admin_manual/configuration_server/oauth2.html
There is also a rumor that said the OpenAI has changed how the model run, now user input is fed into smaller model first, then if the larger model agree with the initial result from the smaller model, then larger model will continue the calculation passed from the smaller model, which supposedly can cut down GPU time.
I use keycloak. Pretty steep learning curve, but once properly set up, it can do pretty much anything.
But if you’re in a pinch, NextCloud can act as an OIDC auth provider out of the box.
Does ssh works? If yes, you can use autossh
to create persistent ssh tunnel to your VPS.
Who cares if it already exists, just make it.
Also consider the possibility when the other, more popular projects got enshittified. Now the fleeing users have an option to switch to your project. It actually happened on one of my side project. I made it because I want to try building my own version of X. It got ~2000 users, but later down the road, X got sold to a new shitty owner that waste no time to enshittify it, and my side project suddenly grow to 20,000 users overnight.