Danilo :exherbo: boosted
Danilo :exherbo: boosted

Plausible Analytics has just hit $2,000 MRR from 300 paying subscribers!

Thank you all for your support, trust and for helping spread the word about Plausible Analytics! 👏 🙏 😍

Danilo :exherbo: boosted
Danilo :exherbo: boosted

C++ standard library, prior to C++20, is unusable. I just hope Clang and GCC hurry up and finish their implementation soon.

Danilo :exherbo: boosted

It is modular in the sense that can you disable module from it. But you can't remove systemd from a system without recompiling half of the packages, which was possible prior to it.

It embeds udev and logind, that are are perfectly fine as standalone libraries. It forces this choice on the user and that's why I think it deserves it.

Regarding the kernel resources, I think those are low-level userspace resources.

Danilo :exherbo: boosted

It looks like Jitsi thinks it's okay to spy/track their users via third party services (like Firebase, Crashlytics, Amplitude) in the name of 'chasing down malfunctions in software'.



I didn't know systemd init tracked kernel resources. Why does it do that?

I think systemd deserve the villification it gets. It came to a modular system and destroyed the modularity. However, the init/service manager is objectively good.

Danilo :exherbo: boosted
Danilo :exherbo: boosted

Zettlr 1.7 FOSS markdown editor for personal knowledge management and publishing

(submitted by DerWOK)

Clang and GCC are perfeclty balanced.

The moment you fix an error in Clang, you got one in GCC.


I think AppImage should be used only with proprietary applications that can't be installed otherwise.

All this new layer and stuff honestly scares me. But I don't wanna say something you have already heard so I'll stop here.

Finally SonarCloud coverage works. I can't tell how much I am happy after countless hours trying to get it working.

Danilo :exherbo: boosted

Here's how it works on sourcehut:

You pay us the subscription fee of your choice and you get access to our services. We spend it on maintaining and upgrading the hardware, and paying for staff to handle development, operations, support, and on-call.

Because we don't take money from investors, we're only accountable to users, and if we aren't working in your interests, we lose our revenue. We are incentivized to work for you, not to generate a return.

Because we're running 100% free software, we're held accountable because anyone can take our shit and run it on their own servers, or even bootstrap a competing company using our own software. We've made it easy to run out out of business if we lose your trust.


Yea, that's true, it can be used anywhere and it is one big advantage of OpenPGP.

Danilo :exherbo: boosted

Completed the provisional procedures for all postgresql operational maintenance:


Still need to test some of these and come up with a monitoring plan, but it's looking pretty good so far. Should be able to completely eliminate total service outages and reduce database-related planned to as little as 30 minutes of read-only service per year

Danilo :exherbo: boosted

I've now opened my own website stats so they are viewable by anyone. Now I just need to start writing some posts and get some traffic!


Danilo :exherbo: boosted

Yeah, I mean they didn’t prove they’re better because they don’t use OpenPGP. It was a random collection of needles into PGP. If Tutanota had forward-secrecy or post-quantum algos it would be a different talk.

> Do you have resource on this? I'd like to research more into this topic.

I can give you a tl;dr version: crypto agility is dangerous, the standard could be slimmed down to avoid some pitfalls (like exabyte chunk sizes) and the working group is basically ran by one person. Browsers had the resources to overhaul TLS so that 1.3 is vastly different and more powerful than what we had years ago. (Not to mention things like Certificate Transparency).

OpenPGP ecosystem participants are doing what they can to improve the status quo but it’s not the same order of magnitude so everything takes longer.


I agree, the part about Tutanota marketing is bad. However some points are valid, in my opinion.

I think that Tutanota is as bad as Protonmail as email standard. I have just moved out of Protonmail into a plain IMAP/SMTP email provider and I am happier now than I was with Protonmail and their lock-in.

> PGP has flaws, definitely, but you won’t know about real problems from this marketing post.

Do you have resource on this? I'd like to research more into this topic.

Show more

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.