"It added an additional three-kilometer-long ring between the main interferometer and the detector to circulate the light before it exited the interferometer’s arms."
Isn't that a delay line? The benefit being that when the undelayed and delayed signals are mixed, the phase shift you're looking for is amplified.
The big players in the Wi-Fi SOC market are all using ARM in recent designs, Intel and its multi-core Atom being an exception. RISC-V isn't supplanting these at the moment. The chipsets in this market are highly specialized: most of the network functionality is offloaded to proprietary hardware peripherals, and the CPU is mostly only doing control plane and bookkeeping. Given the low margins of these products, some designers may eventually drop ARM for RISC-V, but there is a lot of inertia: designers are far more concerned with keeping up with fast-changing wireless standards and delivering new features. The marginal cost saved replacing these small, slow, low-cost, low-power cores isn't a big priority, especially given any risk of delaying new products.
I suspect there is plateau coming in the foreseeable future, as all the most desirable spectrum is fully utilized by maturing chipset designs. Should that happen, cost reduction will become a higher priority.
oh that's interesting. especially the backports. i did not consider that as realistic since perl6/raku is using an entirely new virtual machine. so maybe they are actually reimplementing ideas that are in raku but work with perl 5?
I can't speak to the degree to which Raku VM ideas are being backported. I'm not that deep into Perl. I just know that the Perl 5 language itself is still regularly enhanced, and some amount (can't give you a percentage or speak to the pedigree of all the changes,) of that is inspired by Perl 6. Perl 5 now is a great deal nicer than it was, say 10 years ago, due to these updates.
Sub signatures, defer blocks, official "class", "say", try/catch, enhanced dereference syntax, new and enhanced operators, etc. Significant language changes have been and continue appearing.
It's tragic, and it bothers me that Jens Axboe's work is suffering due to this. Obviously, with the clarity of hindsight, this might have been avoided. Now the cost of the damage it high.
My idea: consider formal verification. A rigorous formal proof of behavior is capable of solving actual flaws and probably capable of overcoming the bad PR.
That would require a great and ongoing effort. However, given the nature of io_uring, it's likely rather amenable to formal verification, and ultimately it is probably necessary. Perhaps our new "AI" tools could greatly reduce the effort.
io_uring is only a more ambitious manifestation of a pattern that has been commonplace for a long time. High performance network, storage and other hardware use circular buffers internally in their drivers to process commands and return results. io_uring is attempting to realize the same pattern, but for the entire operating system interface.
That's a really big idea. It has me asking whether this isn't how all "IPC" should be done, where one generalizes all of the running processes, including the "kernel." Ultimately it has significant implications for CPU architecture, supplanting traditional patterns.
When I wrote "given the nature of io_uring," I had something specific in mind wrt formal verification. io_uring fundamentally consumes a stream of bytes. At the front end of that is a great deal of validation code to ensure the steam contents are sane. At the backend, the commands are mapped to existing kernel functionality. It's the front end I think could be well served by formal verification.
> Those old HPs were engineering workstations built by an HP that at the time still cared about engineering.
I agree. They are heavy and very strong, and chassis is very stiff. That mattered given where some of these ended up being used and how they were treated, despite the cost.
I built a disaster recovery system using python and borg. It snapshots 51 block devices on a SAN and then uses borg to backup 71 file systems from these snapshots. The entire data set is then synced to S3. And yes, I've tested the result in a offsite: recovering files systems to entirely different block storage and booting VMs, so I'm confident that it would work if necessary, although not terribly quickly, because the recovery automation is complex and incomplete.
I can't share it. But if you contemplate such a thing, it is possible, and the result is extremely low cost. Borg is pretty awesome.
I've had one or more around most of my life. They're fun to play with and comedic. They genuinely like their people. They're simultaneously willful and cooperative. They have many habits and preferences, and each one is distinct. They're highly communicative if you understand their motives and language.
If you need to control rodents they are extremely effective and earn their keep. There is nothing more endearing than a proud cat eagerly bringing its catch home to share with its pride. Some people are freaked out by this, not realizing that there is no higher praise a cat can express, hunting on your behalf.
So many dimensions. If a cat likes you it actually likes you: there is no lie in them.
Cats all do the same things. But they have very clearly defined unique personalities, with big variations in intelligence and other qualities.
Also, relaxing. Having a dependent of any kind sleeping soundly close to you implies safety and reassurance in a very primal and satisfying way.
And cats sleep far more than humans do. Even when they're not active they're nice to have around.
I'm not convinced by the "tamed predator" hypothesis. I think if it were true we'd consider them exciting but stressful - like crime fiction.
Clearly we don't. No one sleeps next to a violent crime novel for relaxation.
In fact cat owners often get cognitive dissonance when Furry McPurrFace goes out and eviscerates a bird for breakfast. We feel sorry for the bird, but we don't seriously think "That could have been me. Or one of the kids."
Isn't that a delay line? The benefit being that when the undelayed and delayed signals are mixed, the phase shift you're looking for is amplified.
reply