The article requires logging in to read.

Archived version: https://archive.is/oQ8lm

Just the things others contributed. Unless they somehow get permission from all of their contributors, which, in this case, seems less than likely.

You seem to be slightly confused, that’s how Orkish aviation works. Western FreedomPlanes™ generate lift by harnessing the power of the line that must always go up. For example, the U.S. Dollar’s line has started to go down, which has caused many U.S. airplanes to crash.

After a quick look, looks like it tries to split the (unencrypted) hostname into multiple packets, or at least scramble it slightly. I’m not sure how much it helps in practice, but it might help against naïve filtering/scanning, as the hostname is either sent in different packets, or split and sent unordered in the same packet. It probably only helps if encrypted client hello isn’t supported.

TL;DR: If I’ve understood everything correctly, it just moves chunks of the plaintext hostname around & tries to split it into multiple packets.

Note: Mostly based on comments, as it’s late & I’m too tired to parse too much cryptography code.

Full source of the limit_chunks function, formatted with Rustfmt:

const fn limit_chunks<'a>(
    left: (u64, &'a [u8]),
    right: (u64, &'a [u8]),
    limit: usize,
) -> ((u64, &'a [u8]), (u64, &'a [u8])) {
    let (left_offset, mut left) = left;
    let (mut right_offset, mut right) = right;
    if left.len() + right.len() <= limit {
        // Nothing to do. Both chunks will fit into one packet, meaning the SNI isn't spread
        // over multiple packets. But at least it's in two unordered CRYPTO frames.
    } else if left.len() <= limit {
        // `left` is short enough to fit into this packet. So send from the *end*
        // of `right`, so that the second half of the SNI is in another packet.
        let right_len = right.len() + left.len() - limit;
        right_offset += right_len as u64;
        (_, right) = right.split_at(right_len);
    } else if right.len() <= limit {
        // `right` is short enough to fit into this packet. So only send a part of `left`.
        // The SNI begins at the end of `left`, so send the beginnig of it in this packet.
        (left, _) = left.split_at(limit - right.len());
    } else {
        // Both chunks are too long to fit into one packet. Just send a part of each.
        (left, _) = left.split_at(limit / 2);
        (right, _) = right.split_at(limit / 2);
    }
    ((left_offset, left), (right_offset, right))
}

Same, but for write_chunk:

fn write_chunk<B: Buffer>(
    offset: u64,
    data: &[u8],
    builder: &mut packet::Builder<B>,
) -> Option<(u64, usize)> {
    let mut header_len = 1 + Encoder::varint_len(offset) + 1;

    // Don't bother if there isn't room for the header and some data.
    if builder.remaining() < header_len + 1 {
        return None;
    }
    // Calculate length of data based on the minimum of:
    // - available data
    // - remaining space, less the header, which counts only one byte for the length at
    //   first to avoid underestimating length
    let length = min(data.len(), builder.remaining() - header_len);
    header_len += Encoder::varint_len(u64::try_from(length).expect("usize fits in u64")) - 1;
    let length = min(data.len(), builder.remaining() - header_len);

    builder.encode_varint(FrameType::Crypto);
    builder.encode_varint(offset);
    builder.encode_vvec(&data[..length]);
    Some((offset, length))
}

Link to the MIT license file

Is silicone rigid enough? Adult toys have slightly different requirements from many other 3D printed things.

How do you usually do that?

Do you create a negative of the model digitally or take a mold of the printed parts?

What kinds of human-friendly materials do you use? And do you need any special equipment for it?

With the new Swap™ technology, you are no longer* limited** by your RAM. Our*** brand-new**** Swap™ technology turns your unused disk space into usable***** memory at almost****** no******* perceivable performance impact. When combined with our ZSwap™ compression technology, you can now achieve an up to 5000% or better******** unused-disk-to-memory conversion ratio********* than many RAM-downloading services.**********

Isn’t that how it’s done in English (Traditional)?

I hope that the language’s ints are at most 32 bits. For 8 bits it could even be written by hand & the source code for a 32 bit version would only take up avg_line_len * 4GiB space for the source code of the function. But it might take a bit of time to compile a version that supports the full range of 64 or 128 bit ints.

So it works now? If so, then glad to be of help.

Just remember that if your shebang points to sh, you can’t rely on bash-specific features. The shebang line basically tells the kernel to run your file with the specified program. So, for example, a file with #!/bin/cat will print the full contents of the file (including the shebang) and #!/bin/echo will print the command line. (something like ./script arg1 arg2) As the echo command does not try to interpret arguments as paths, the content of the script would be ignored in that case.

Does the script file have a shebang? (Something like #!/bin/sh as the first line)

In that case, don’t look up Dwarf Fortress. I’ve also heard good things about Factorio, but I haven’t tried it yet.

And C++ compile times made me learn C. You must be jealous that my memory errors & race conditions compile faster than yours.

I haven’t gotten a chance yet, as rustc is using 100% CPU & RAM.

Nice!

I’ve read that Xlibre is a fork of Xorg that would still get new features, so I did not know much about it. I use Wayland, as it works better for me, but after reading the readme, I somehow want to use it even less.

This fork was necessary since toxic elements within Xorg projects, moles from BigTech, are boycotting any substantial work on Xorg, in order to destroy the project, to eliminate competition of their own products. Classic “embrace, extend, extinguish” tactics.

How is it EEE? A point about embracing could be made, but wouldn’t there have to be extension with non-FOSS code or difficult to implement additions to standards? Even then, as it’s hosted on GitHub of all places, it doesn’t seem to be that extinguished.

Right after journalists first began covering the planned fork Xlibre, on June 6th 2025, Redhat employees started a purge on the Xlibre founder’s GitLab account on freedesktop.org: deleted the git repo, tickets, merge requests, etc, and so fired the shot that the whole world heard.

I haven’t heard about this, but I would guess that it’s from a CoC violation, based on the rest of the readme.

This is an independent project, not at all affiliated with BigTech or any of their subsidiaries or tax evasion tools, nor any political activists groups, state actors, etc. It’s explicitly free of any “DEI” or similar discriminatory policies. Anybody who’s treating others nicely is welcomed.

Calling DEI discriminatory is never a good sign.

Together we’ll make X great again!

This seems pretty close to a certain right-wing slogan from the USA. I don’t know if it’s accidental or a joke, but I don’t think that it’s appropriate, and like the author, I don’t have much good faith left for this.

A lot of commits by the Xlibre developer were reverted upstream and the readme does not look professional.

They’re on the feddit.uk instance, so it’s more likely that they’re British. Either their VPN prefers USA servers for some reason, or they’ve only done about 138 million downloads at most.

What’s disgusting about it? The only thing I can think of is the implicit return, which felt a bit icky at first.

Also, as the if expression is an expression, you can call methods on it like so:

if 1 > 2 {
    3
} else {
    4
}.min(5)

(the above is still an expression, so it could be used, for example, as part of a condition for another if)

Of course, you can write horrible code in any language, but the ability to use blocks where expressions are expected can be great sometimes.

Rust would allow you to

let ret = if some_condition {
    <a lot of expensive calculations>
    result_of_operations
} else {
    <a lot of other different expensive calculations>
    result_of_other_operations
};

Now you don’t have to declare it inside the blocks.

It’s way more common than you may realize. Intel & AMD (and other x86 CPU manufacturers of the time) did it before the first Crusoe CPU launched. (2000 according to Wikipedia)

CISC architectures are now seen as inefficient, so all the new ones are RISC and new CISC CPUs just translate the instructions to their intenal RISCier microarchitecture. The CPU’s microcode specifies what an instruction translates to.