jdub

That's not how trademarks work.

The Tolkien state (and related orgs) have registered "Palantir" under many (generally publishing/merchandising) domains, but Palantir Technologies has registered it under IC 009: computer software for collection, editing, analysis, viewing, [etc.] and a few others.

Calling it "Harmony" is a historical death knell. :-D

What makes you believe that most things have been digitised in the first place?

Young singers brought up listening to autotuned vocals can unknowingly learn and emulate the sonic signature of the tuning algorithm (and the telltale lilt when it's used as an effect, but the subtle tuning case is more surprising).

If you read too much sloppy LLM prose, it's going to influence how you write and structure your own.

The USA didn't New Deal hard enough to keep up with the rest of the "Western" world's more humane form of capitalism, which you can see in how consistently the USA lags behind other countries in measures of actual human progress and comfort.

While most "Western" countries suffered through similar technocratic, neoliberal turns in the 80s, they were built on a stronger social democratic base than the USA.

And don't get fooled by "horseshoe theory" ignorance: Fascist states were/are authoritarian by definition. Communist states were/are authoritarian by culture.

Before OLED (and similar), most displays were lit with LEDs (behind or around the screen, through a diffuser, then through liquid crystals) which was indeed the dominant power draw... like 90% or so!

But the article is about an OLED display, so the pixels themselves are emitting light.

Even at the best of times, the United States does not have a strong culture of observing or learning from other nations.

> At Reco, we have a policy engine that evaluates JSONata expressions against every message in our data pipeline - billions of events, on thousands of distinct expressions.

The original architecture choice and price almost gave me a brain aneurysm, but the "build it with AI" solution is also under-considered.

This looks like a perfect candidate for existing, high quality, high performance, production grade solutions such quamina (independent successor to aws/event-ruler, and ancestor to quamina-rs).

There's going to be a lot of "we were doing something stupid and we solved it by doing something stupid with AI [LLM code]" in our near future. :-|

No, it pays for ticket-clipping middle-men and political corruption.

Or, the ability to construct additional sentences influenced by prior ones.

Digital neural networks and "neurons" were already vastly simpler than biological neural networks and neurons... and getting to transformers involved optimisations that took us even further away from biomimicry.

This is a very optimistic, pro-technology-cleverness point of view.

I recommend reading the linked persona selection model document. It's Anthropic through and through - enthusiastic while embracing uncertainty - but ultimately lots of rationalisation for (what others believe is) dangerous obfuscation.

But there's a spectrum of responses to these technologies, from knee-jerk cynicism to genuine moral disgust. "Useful" and "good for people/society/humanity" don't always go hand-in-hand, particularly if you take origins and power into account.

You're likely to find more nuance in opposing views than your "underwhelmed by AI" generalisation could represent.

Did you see the network security stock sell-off after Anthropic announced a code security analysis feature? There's a sliver of nothing between mob mentality and wisdom of the crowd.

How would copyright law possibly compel the burning of books?

> Netscape didn't attack Microsoft's business software, operating systems or other pieces of their offerings.

That's not how it was interpreted at the time: Netscape threatened to route around the desktop operating system (Win32) to deliver applications via the browser. "Over the top" as they say in television land.

Netscape didn't succeed, but that's precisely what happened (along with the separate incursion of mobile platforms, spearheaded by Apple... followed quickly by Google, who realised they had to follow suit very quickly).

> And neither had the capability to build large software very fast.

Internet Explorer. Android. Gemini.

History suggests otherwise, and there's nothing particularly special about this moment.

Microsoft survived (and even, for a little while, dominated) after missing the web. Netscape didn't eat its lunch.

Then Google broke out on a completely different front.

Now there's billions of dollars of investment in "AI", hoping to break out like the next Google... while competing directly with Google.

(This is why we should be more ambitious about constraining large companies and billionaires.)

It's so refreshing to discover that the "I found one bug in systemd which invalidates everything" pattern continues in the year of our lord 2026.

Sure, and systemd would translate that directly into a dependency on network startup, which is precisely equivalent to the approach I mentioned that depends on operator knowledge. It's configuration, not "just works" inference.

Sounds like your admin, distro, or the systemd team could pay some attention to systemd-update-done.service

The "can only be used in /etc/fstab" systemd settings are essentially workarounds to do those things via fstab (and workaround fstab related issues) rather than depend on other systemd facilities (c.f. systemd-gpt-auto-generator). From a "what can you do in /etc/fstab without knowing systemd is working behind the scenes" point of view, then yes, systemd units are vastly more configurable.

> After 5 minutes I got freshly burned floppy.

oh god

OK, think it through...

How do we determine that a specific instance of a filesystem mount is "remote", or even requires a "network"? Consider that the network endpoint might be localhost, a netlink/unix/other socket, or, say, an IP address of the virtual host (practically guaranteed to be there and not truly "remote").

systemd has .mount units which are way more configurable than /etc/fstab lines, so they'd let you, as the administrator, describe the network dependency for that specific instance.

But what if all we have is the filesystem type (e.g. if someone used mount or /etc/fstab)?

Linux doesn't tell us that the filesystem type is a network filesystem. Linux doesn't tell us that the specific mount request for that filesystem type will depend on the "network". Linux doesn't tell us that the specific mount request for that filesystem type will require true network connectivity beyond the machine itself.

So, before/without investing in a long-winded and potentially controversial improvement to Linux, we're stuck with heuristics. And systemd's chosen heuristic is pretty reasonable - match against a list of filesystem types that probably require network connectivity.

If you think that's stupid, how would you solve it?

LFS never had academic, educational, or pedagogical merit. It was always sheer faith that by doing enough busywork (except the truly difficult stuff), something might rub off. Maybe you learn some names of component parts. Components change.

Weeeeelllll...

Post-CFEngine (Puppet, Ansible, Terraform) and cloud platform (CloudFormation) infrastructure-as-code is over a decade old.

Docker's popularisation of containers is just over a decade old.

But containers (and especially container orchestration, i.e. Kubernetes) are still entirely ignorable in production. :-D

I'm not threatened by the idea that LLMs might actually be intelligent. I know they're not.

I'm threatened by other people wrongly believing that LLMs possess elements of intelligence that they simply do not.

Anthropomorphosis of LLMs is easy, seductive, and wrong. And therefore dangerous.

Why do people even begin to believe that a large language model can usefully understand and interpret health data?

Sure, LLM companies and proponents bear responsibility for the positioning of LLM tools, and particularly their presentation as chat bots.

But from a systems point of view, it's hard to ignore the inequity and inconvenience of the US health system driving people to unrealistic alternatives.

(I wonder if anyone's gathering comparable stats on "Doctor LLM" interactions in different countries... there were some interesting ones that showed how "Doctor Google" was more of a problem in the US than elsewhere.)

Why do you have those expectations?

> But now that most code is written by LLMs...

Pause for a moment and think through a realistic estimation of the numbers and proportions involved.

(Ha, nice to see Jon Corbet's name on the PCI Endpoint documentation...)