2·
3 days agosame or less memory than the previous generation
means less room for bloatware
moved from us instance
- 0 Posts
- 54 Comments
Joined 5 months ago
Cake day: June 3rd, 2026
You are not logged in. If you use a Fediverse account that is able to follow users, you can follow this user.
No because there’s the Steam update .sh script. AKA voluntary RCE.
the only good thing to come out of the ai bubble
Still a browser. Abaddon is nice but no feature parity (especially compared to browser + equicord, I am using like 1 plugin from there rn).
HOLY FUCK
This person on the internet gives a shit! Going back to your webshit browser only adds .25 seconds on tiling and .5 on floating. So much time to do useful stuff, like playing your games with strangers on the interwebs!
Also
Linux didn’t have the most basic voice chat function up to now??? :O
WHO the actual FUCK submitted voice chat to the Linux mailing list? Was it some
@redhat.comguy again? Or was it some LLM owned by Broadcom? Torvalds accepts anything at this point, he got the AI psychosis. Oh, you are talking about Discord, and specifically the .deb files from their website people fuck only to have another copy of Chromium? Holy fucking shit.
“fork” implies there’s code to fork
Wow. Amazing! I use it in a browser when I need it then close the tab when I don’t, you know, like a reasonable person? I am not installing more Chromium to not only spy on me but literally also tell everyone what I am doing unless I go out of my way to opt out (the toggle will just turn back on whenever it feels like it).
I bet they did some dumb shit like subscribe every single QWERTY 101 key.
3·7 days agoYou have ignored the other suggested project in that comment.
Doing bad decisions is part of simple sabotage, comrade.
4·11 days agoType
man 3 open.That’s what most Flatpak apps may do instead of asking permission via a fancy menu, which actually just does the same thing.
Hmmmmmm, maybe then the virus shouldn’t use GTK dialogs for opening files and call open(3)? hmmmmmmmmmmmmmmmmmm
Or
--break-system-packages, then never run as root and runrm ~/.local/lib/python*to clear.
I can help.
First, install it. This depends on your distribution and package manager.
apt,pacman, you name it, just search for “yt-dlp”. Epic tech tip: do not install from flatpak or snap, because if you ever run into an issue the maintainers will hate you for it, and because it will be annoying (slightly different) to use in the former case. Important to note that sometimes your distribution will put some technically-optional component (that you probably still want) into a different package likeyt-dlp-ejs. Regardless, you will probably want to installffmpeg,curlandcurl-cffi(sometimes calledcurl-impersonate) for all of the features.If your distribution allows using
pip(most don’t), you may dopython3 -m pip install -U "yt-dlp[default]"to install it (do the same to upgrade later).Navigate your shell to the directory you want to put your video in (some file managers have “Open in {some terminal emulator}” in the right click menu, which does as it sounds).
Now,
yt-dlp --embed-metadata -- video_id some_other_video_id(--embed-metadatais optional, it’s so that someone who downloads the video will be able to read the description). Note the--, it tells yt-dlp that your options have ended and IDs have started, to prevent it from thinking an ID is an option. In extremely rare cases, the video ID may have an exclamation mark or an ampersand, which will fuck everything up. In that case, put \ in front of the symbol, right in the ID (like “dwQ&smshIt” turns into “dwQ\&smshIt”). If you are downloading from a service other than YouTube, paste your URL (with the same precautions) instead.Welp, this is the basic usage. You can stop reading if it already did what you want.
This may sometimes save to an MKV file when it’s impossible for it to be WEBM, which is good for video editors. Most web browsers don’t understand it, however. So, if you want to share it online, you may want to instead do:
yt-dlp --remux-video mp4 -- video_idto forcefully shove it up a.mp4file. If you have only noticed later, you can also do:ffmpeg -i "video.mkv" -c copy -movflags +faststart "video.mp4"(and delete the old one).This picks and downloads the highest quality format it detects. You can see the available formats for a given upload with
yt-dlp -F -- video_id. It is a table with format numbers, video resolutions (and if HDR is present or not), audio frequencies and so on. Note that most only have video OR audio, but some have both. For YouTube, the objectively best audio is 251, and is present in basically all videos (and music!) since 2014. Available video depends; apart from resolution there’s also the video codec. My ranking from best quality per filesize to least resource heavy to decode (for old phone users…): AV1, VP9, VP8, H264.Now, you can specify the formats you want to put into your video like:
yt-dlp -f "251+699" -- video_id. That’s it. If you want to download from YouTube Music without the pointless video tracks, you may also do:yt-dlp -f 251 --embed-metadata --remux-video opus -- your-amazing-id-here.Downloading age and/or account gated content, and content from some picky sites, is a little more tricky. Login to the service from Firefox, default profile and no containers (it’s possible to use other profiles and containers but the syntax is annoying and you will have to dig around directories). Now, do
yt-dlp --embed-metadata --cookies-from-browser firefox -- id. This will download using your account. Although this may also make the site refuse some formats and give you a piece of shit video. If it’s not doing anything still, try adding--impersonate firefox. If you are really desperate, you may try installing Chromium, then logging in there and using--cookies-from-browser chromium --impersonate chrome.Bonus tech tip: there’s some basic support for the Wayback Machine. Try:
yt-dlp --embed-metadata -- https://web.archive.org/web/0/https://youtube.com/watch?v=wvsboPUjrGc https://web.archive.org/web/0/https://youtube.com/watch?v=_jZXbhe6-rg.
¯\_(ツ)_/¯*
Also, this specific example isn’t ASCII art, it’s some other encoding which I forgot the name of, Japanese characters, variable width (in bytes). Common for it to be drawn using Arial or a metric compatible font rather than any monospace.
Note that “ASCII” art also refers to 8-bit extensions of the encoding, where “oldskool ASCII art” means strictly the actual 7-bit set. Not to be confused with ANSI art, which is the same as ASCII but with ANSI control characters, like for changing color.
“text art” is what I usually call everything combined, including art on the Commodore systems, only very recently made possible with Unicode officially. Anyways, I don’t really see any of it on the image…
2000s != “ASCII-era”. But cool regardless.
There are also other socialist nerds who have decided they don’t want to obligate the corpos to give them code because it sucks. Welcome to BSD.
Not while all of the memory is going to the datacenters. Commercial applications may be built on some Raizen 407000, which sucks, but hobbyist projects will all be made on regular computer, during the shortage. Unfortunately, Linux isn’t really a hobbyist project… Was thinking NetBSD.