Like every other *online* digital assistant.

The offline ones just haven't been made yet.

Scanned photos of real-life toy guns and sculpted clay and Latex Demon models perhaps?

Start over again, and then a browser will rise from the ashes like a ... Phoenix?

AI upscaling and denoising works best on line art/anime art, because it's much simpler than actual photos, so it's easier to predict how the lines and curves will extend.

TAR is not a compression tool, it is a tool to stream a series of files to another stream. Capturing that stream to a file just happens to be one possible way to use TAR. It can also be used as a hardlink-preserving file copy command when used with pipes. Can even pipe it through netcat and now it's a way to transfer files over a network.

But it's horrible for creating archive files. No seeking support, need to stream through the whole archive file (possibly skipping through the data if the TAR file is not compressed) to even enumerate the contents.

Maybe you want to back up files and maintain the hardlinks.

Or maybe you want to know if writing to one file will modify another file.

Your tool for creating hardlinks is a userspace tool.

Isn't a filesystem-wide unique Inode ID required to implement hardlinks?

As in, any two files with the same Inode ID, size, and a hard link count greater than 1 must be hardlinks of each other if they are on the same mounted partition.

I love these images, it's a good test case for your content-aware fill tool/inpainting method of choice.

DOOM did not run on a pregnancy test. That was a case mod where a different screen was stuffed into the shell.

The downside to using a real domain is the chance of a misconfigured system making everything public instead of keeping it internal.
The downside to using an arbitrary domain is that someone could register the top-level domain, and then we're back to the possible misconfigured case again.

Tar still has special properties that come from it not supporting any form of seeking. You can use Tar with pipes, including over netcat. And it will even replicate your hardlinks and symbolic links.

19GB TAR archive is the worst possible use of Tar. You'd want to use literally anything but Tar for this use case.

Why? Tar provides no random access at all, the entire file needs to be streamed through in order to list what files are in the archive.

If it's Microsoft's servers, and users need a Microsoft account to access this, then Taylor could sue to try to force Microsoft to disclose whose Microsoft accounts have been generating those images.

Not your computer, not your privacy.

You don't want 4K on a handheld screen, you want 4K on the TV. You want the handheld to have have the best possible battery life, and a lower-resolution screen is how you do that.

Startpage gives me captchas on searches for some reason, don't like using them.