Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
goombacloud
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
goombacloud
7mo ago
They won't build software, they'll let some AI-based software do the execution of their instructions (which is inefficient, opaque, vendor-locked, not reproducible etc.)
2.
▲
by
goombacloud
2y ago
If you mean using Incus on Flatcar, there is a PR for adding Incus as systemd-sysext extension. Flatcar inside Incus is a bit more difficult: for Flatcar being a container one can import https://stable.release.flatcar-linux.net&#
3.
▲
by
goombacloud
2y ago
With static binaries that is not needed (and you can use OS=_any in the extension release file to mark them compatible). If you want to repackage distro binaries without recompilation, you can have a look here: https://github.com
4.
▲
Flatcar: OS Innovation with Systemd-Sysext
(flatcar.org)
120 points
by
goombacloud
2y ago
|
18 comments
5.
▲
by
goombacloud
2y ago
In https://archive.softwareheritage.org/browse/revision/e446ab7... one can open the patches and then click the "Changes" sub-tab. Stuff like this looks like a perf improvement but who knows if a tricky b
6.
▲
by
goombacloud
2y ago
This might not be complete because this statement "More patches that seem (even in retrospect) to be fine follow." lacks some more backing facts. There were more patches before the SSH backdoor, e.g.: "Lasse Collin has alread
7.
▲
by
goombacloud
3y ago
When socat is around a simple server can also be constructed with it: tee /tmp/server > /dev/null <<'EOF' #!/bin/bash set -euo pipefail SERVE="$1"
8.
▲
by
goombacloud
3y ago
I really think we should have means of spawning wasm components from wasm components. How the runtime runs them should be up to the runtime - it could be directly backed by kernel primitives but it could also be in a browser. Leaking posix
9.
▲
by
goombacloud
3y ago
To spot more common problems I recommend: alias shellcheck='shellcheck -o all -e SC2292 -e SC2250'
10.
▲
by
goombacloud
3y ago
For regular Linux users you can do: sudo touch /etc/udisks2/tcrypt.conf sudo systemctl restart udisks2 and then any veracrypt volumes can be used in Nautilus or GNOME Disks similar to LUKS volumes.
11.
▲
by
goombacloud
3y ago
Does this finally have UEFI by default?
12.
▲
by
goombacloud
3y ago
The next logical step would be to only support wasm programs ;)
13.
▲
by
goombacloud
3y ago
gzip uses LZ and Huffman coding and not arithmetic coding with a predictor, so yes, these are not similar.
14.
▲
by
goombacloud
3y ago
Specially compression algos that use arithmetic coding with interval weights adjusted based on the prediction of what is likely coming next are very similar. They adjust the arithmetic coding ( https://en.wikipedia.org/wiki&#
15.
▲
by
goombacloud
3y ago
Audio is covered as well, as written in the article, I recommend reading it.
16.
▲
by
goombacloud
3y ago
https://www.mojeek.com/ even has its own crawler
17.
▲
by
goombacloud
3y ago
There are multiple ways to start containers, one of them is to define systemd units in Ignition. If that part needs to change, reprovisioning is needed but a new "flatcar-reset" tool (in Alpha) should make the barrier for reprovis
18.
▲
MNT Pocket Reform – A Technical Walkthrough
(crowdsupply.com)
1 points
by
goombacloud
3y ago
|
0 comments
19.
▲
by
goombacloud
3y ago
There is a changed version that includes drop-in config files, masking, and loading defaults from /usr/ instead of requiring full config files to live under /etc: https://uapi-group.org/specifications/spe
20.
▲
by
goombacloud
4y ago
For most use cases you can configure autosuspend and always wrap your commands with systemd-inhibit or gnome-session-inhibit. E.g., for an SSH session when starting tmux, or when you start the music player (for GNOME I guess that many playe
21.
▲
by
goombacloud
4y ago
You can spawn a (priv) container with the host PID namespace and then trace from there.
22.
▲
by
goombacloud
4y ago
Not really, you can start it from the host and trace any container process you like. On Linux the processes are in one large tree and the host has IDs for all container processes. Inside the container you may have different IDs if a pid nam
23.
▲
by
goombacloud
4y ago
You don't need to be in the container and trace from there, you can run strace outside of the container (you just have to make sure that the process ID you use is the one as seen from the host).
24.
▲
by
goombacloud
4y ago
I find myself using cat or less even thought I have bat around because copying from cat or less works best and you don't get line breaks inserted.
25.
▲
by
goombacloud
4y ago
It's sad to see that Apple Silicon gets extra treatment and arm64 work suddenly gets done while for Linux it's still amd64-only. With Signal it's the same and they also don't offer arm64 builds. Nowadays it's very e
26.
▲
by
goombacloud
4y ago
I think the hard part is to know when caching is not allowed, and for that the compiler is a good point to manage such a cache because it could add metadata such as compiler version (maybe even patches?), compiler flags etc which a wrapper
27.
▲
by
goombacloud
4y ago
Building into the kernel instead of using modules is not perfect because it increases RAM usage when you don't use certain modules compared to having them loaded on demand.
28.
▲
by
goombacloud
4y ago
Love this modularity of swapping the compute modules!
29.
▲
by
goombacloud
4y ago
By the way, the leading nncp in the LTCB (text.html) "is a free, experimental file compressor by Fabrice Bellard, released May 8, 2019" :)
30.
▲
by
goombacloud
4y ago
The comparison doesn't make much sense because for fair comparisons you have to measure decompressor size plus encoded image size. The decompressor here is super huge because it includes the whole AI model. Also, everyone needs to have
More ›