I’m not sure why people keep pushing that myth on C++. It’s been a decade we have smart pointers. There’s no memory management to be done ever.
Using the old ‘new’ is like typing ‘unsafe’ in rust. Even arrays/vectors have safe accessor.
Am I missing something?
Are we codeberg yet?
Thank you for sharing this. I didn’t know this FS yet. It seems new and have some nice goals. I always have a grudge against zfs/btrfs because of the resource usage/performance.
I’ll keep an eye on this. I’d love to find some benchmarks.
I am so sorry, I am going to fix this. Also I am not one of the people who downvoted your comment. I like when people point at my mistakes. I am making a lot of those even in my native language.
I tried watching the video. I am genuinely interested but I couldn’t. The video is uncuted (edit: uncut*, my English is so bad) and very slow paced. After 10 minutes I gave up (50 minutes remaining).
Maybe an other time or with an other video.
I thought this also would need the next Nvidia driver version 555. Am I missing something?
it’s been at least 7 years now, I don’t have any hope.
Wine is fine so I guess proton would be.
Damned, this is so frustrating when you cannot switch yet. Not like Wayland is perfect anyway but I felt the same with pipewire where the new system as some needed improvement but the switch is harsh.
They list gitea but not forgejo. That’s not really advocating for FOSS. “all” (the ones I looked at) are startup products coined as open-source.
I really don’t like this website and this list, to me this is replacing bad solutions by other bad solutions (I am sorry for the people that like firebase and co).
I am sorry for the negativity but I really don’t enjoy this link and all it represents and all the people enjoying such content. I guess I/we should explicitly separate FOSS from open-source.
I may be out of touch and should be educated on why/how this is good.
I don’t like it being HTTP based and TLS (certificate?), nor I am a fan of flutter and the other 70-ish dependencies (https://github.com/localsend/localsend/blob/main/app/pubspec.yaml).
KDEConnect is great and does way more than file sharing, I’ll stick with that.
On top of the fact that those previews are annoying as hell as other comments pointed out, I want to add that this kind of feature also uses a fair amount of processing + memory.
I think that is a nice opt-in feature for those who wants it but I like my default light and simple.
Thanks for the link, I knew hyperbola for many years from afar. Reading this gave me a lot more insight on the project. I find it very cool and pushing toward better software like GNU, openBSD and suckless.
Nothing is perfect but for server this distribution could be a nice option . I’d love to see an arm version of it. I guess RISC-V would also be a perfect match for them
It really has the vibe of an hbomberguy video. I also feel the background is a subtle tribute to his style.
They just don’t want to watch it. Length is not really an argument.
Celeron is x86 and quiet bad at its job as far as I have always seen. Benchmark would be nice.
Spending 3000$ to assemble a PC and installing TrueNas as is not new nor informative to me. I’m not a huge fan of rich people spending thousand of dollars building a power hog a call it a nice NAS.
The post would be more interesting if, at least, it would show the assembling (not stolen photos) and the software settings. For a really interesting post I would like to see a before/after benchmark of performance and power consumption.
I don’t think any server, especially self hosting should be CISC based like x86 architecture but RISC like arm or RiSC-V. The power usage is a order of magnitude less. Also really building yourself would probably cost hundreds and not thousands of dollars.
And C++, just checked the wiki and the 2 example of openssh’s heartbleed and sudo, both in C. Not C++. As expected.