kbwt

I work on software for augmented reality and distributed systems. That background is not directly applicable to quantum physics, but I like to think that I am highly trained at spotting logical errors.

So yes, it "tunnels" through the timer field also, but the probability of interacting with the timer field is not zero.

Of course, but it is less than one. Hence the measurement underestimates the proper time interval.

I am not a physicist, but the experiment does not seem convincing to me.

The Larmor clock doesn’t measure a proper time as in spacetime distance between two events. Instead it measures the rotation of a dipole in an electromagnetic field.

The experimenters apply a combination of an electrostatic field (the barrier) and a magnetic field (the timer driving Larmor precession). In the frame of reference of the spinning particle, this is exhibited as a sum of two electrostatic fields. The barrier is a locally uniform repulsive electrostatic field, while the the timer is a radially varying axial electrostatic field. The particle is a dipole, tilted from the timer axis by the precession angle. When the particle tunnels through the barrier, it also tunnels through the timer field, without precessing over the tunneled distance just as the particle is not exhibiting repulsion from the barrier over the same distance.

This is experimentally verifiable as the amount of missed precession has a lower bound proportional to the sine of the angle between the magnetic field and the barrier.

And upgrade your server software within hours of vulnerabilities being published if you want to stay secure: https://www.openwall.com/lists/oss-security/2019/08/28/3

From a few days ago:

* CVE-2019-11500 : Critical Dovecot and Pigeonhole vulnerability (https://www.openwall.com/lists/oss-security/2019/08/28/3)

I still run my own mail server, but I hate having to keep up with these security vulnerabilities which can come up at the most inconvenient time.

I would think exactly the opposite.

In your example, how do I embed a Foo in my FooAndBar object?

On a small scale, planar waves can be modeled like flat sheets of paper traveling through space without any angular momentum (no twisting motion).

They certainly have angular momentum, it just depends on the choice of origin. If you pick an origin along the peak ray of the plane wave, there will be no twist around that point. Just like with a particle traveling in free space.

At that point, what does Protobuf really buy you?

All of those will be around not because they are good, but because so many other things are built against their interfaces.

The application is using a proprietary client/server protocol, so it already lacks lacks any kind of interoperability.

In this specific case, it's unclear whether the bug has direct security implications. The broken SHA-1 is used on some user-controlled data that gets XORed onto the server's decryption of a user-specified payload before being passed into an RC4 key schedule. It's certainly plausible that this might produce a server-assisted privacy compromise of other users' sessions.

Anecdotally I know of one very popular application that has a bug in their SHA-1 implementation causing it to effectively only perform a single round.

In the currently proposed revision, the law applies to anyone facilitating access to an internet-based service "unless the criminal use of the service is of insignificant importance".

See the changes relative to the original draft here: https://www.internet-strafrecht.com/wp-content/plugins/downl...

There is also a huge discrepancy between the proposal from January and the proposal that is being voted on: https://www.internet-strafrecht.com/wp-content/plugins/downl...

A few highlights:

- Instead of "providing an internet-based service", the crime is now "facilitating access to an internet-based service".

- It is no longer a requirement that access to the service be restricted.

- The law now applies to all crimes, rather than just the specific crimes that were listed before.

No, this law was proposed by NRW and Hessen, but if passed it would affect all of Germany.

Latest revision of the law drops the "restricted access" condition.

Diff can be found here: https://www.internet-strafrecht.com/wp-content/plugins/downl...

That is not the version of the law that is being voted on. The amended version is much broader, doing away with the restricted access condition, and now applies to all crimes rather than a specific list. But the worst part, the text has been changed from "provide" to "facilitate access to", meaning this could also apply to ISPs and free software contributors.

Diff from the initial proposal: https://www.internet-strafrecht.com/wp-content/plugins/downl...

I haven't heard of anyone in Europe reporting their salary after-tax.

Simple. You can only get old bonds on the market, where that surplus is priced in.

iOS apps aren't Apple's products.

But then you pay the cost every single time on startup, unless your components barely interact.

Using techniques like ranges [...] _dramatically increases_ optimized performance vs. naive techniques.

This claim will require some evidence. In my experience, it's extremely common for novice engineers to trade orders of magnitude in build time overhead chasing negligible runtime performance improvements.

On big projects, linking times are really the bottleneck.

Sure, incremental linkers exist [1], but all of them tend to do O(n) work on every invocation. Source control software, even developed explicitly to scale [2], behaves the same way. Makefiles as well; Tup [3] tries to solve this, in vain since the linking step is still holding everything up. There is so much inherent inability to scale built into our tools. So on big projects everything grinds to a halt as you cannot buy enough developers and hardware to keep up with O(n^2) forever.

[1] MSVC cl, GNU gold

[2] git: http://lkml.iu.edu/hypermail/linux/kernel/0504.0/2022.html

[3] tup: http://gittup.org/tup/manual.html

Recent article regarding D compile times: https://blog.thecybershadow.net/2018/11/18/d-compilation-is-...

I've personally been watching Rust from the sidelines. I am still waiting for a language to tackle excessive compile and link times.

ICOP is the opposite of polling. I really don't see the resemblance here.

Sure, but this patch is only useful for single-purpose, HFT-like workloads. Batched io_submit on sockets would be useful for any application sending many small packets to lots of clients, such as game servers.

There's io_submit which lets you batch IOs on any number of fds. It appears to lack socket support, but I don't see any reason why that couldn't be added.

I'm relatively sure ICOP still requires two syscalls per IO.

The only thing this really improves is latency, because you can always batch more I/O into a single syscall.