InTheArena

There is a reason why Fentress has not been used since the airport actually opened. All of this stuff that's been done since, including the overhaul of the main terminal, has been done by other architects.

Good articles - there is a fantastic set of books on the engineering lessons from the airport.

A few things the article gets wrong - there were in fact long negotiations before ground was broken - and signed contracts. The CEO is continental was one Frank Lorenzo - and for a variety of reasons he played political games with the FAA and Denver before he was kicked out - and forbidden legally from ever running an airline again.

Part of what also happened is that united passed conditions o designed to cripple Competitors at Denver - going so far as demanding that the bridge between the main terminal and A concourse (where continental and other airlines were located) be frosted to make it less appealing for Continental customers. It’s also the reason why terminal b is physically wider then A or C.

The article doesn't talk about why they built it, really, but the big reason is that the old runways for the airport were so close together that any sort of weather dramatically reduced capacity at the airport (to the point that the old Stapleton Airport at Denver was the number one delayed airport in the nation, a really big problem when there were sometimes between three and five different carriers hubbing there).

A good example of something that the engineers got right - they replaced a heavy steel structure for a glass roof on the main terminal with a tent. (this why concoursesA,C and especially B have glass roofs while the maybe terminal has white canvas. Not only did this end up being a massive cost and weight savings - it gave the airport its iconic shape. Despite planning on a 10-year refurb cycle for the roof, they haven't had to replace it once.

the last vestiges of the old barge system are finally going away - and the tunnels are being reclaimed to provide a walkway to concourse B and c.

Why are they doing that now? Because the airport has actually been a dramatic success as a connecting and as an O&D airport. It now hubs three different airlines and is well over its original design maxes. They've been doing massive construction at the airport and are getting ready to break ground on two new concourses to continue to meet the demand.

Each of these projects are qre dramatically cheaper than competing efforts of other airport with tens of billions of dollars to refurbish extraordinarily constrained terminals and runways. Over the long run, it seems like a pretty good decision to build the airport, even if it got a rough start.

This is exactly correct in the context of the decision as it played out, but stretches and re-engines are very common in commercial airlines. Changing that decision without reinforcing the safety culture probably would have ended in the same place.

no. I am pointing out that a hyperbole from the EU is likely to cause a problem after this dipshit has left the oval office.

Imagine what the EU reaction would be if the United States offered to make bring in the UK or Poland as state(s).

Geopolitical games like this are how real wars eventually start.

This part gives me hope, but I have not been able to find the certificate:

Option B: cloud mode without Developer Mode

Starting with v2.0.0 the plugin can also drive a cloud-paired printer with verification left ON — no Developer Mode. This keeps the cloud features (cloud print dispatch, print history, MakerWorld) but is meant for advanced users, because it requires private slicer credentials that this project does not distribute.

Two things are needed:

1. Bambu's slicer credentials, which you provide yourself. Put slicer_cert.pem, slicer_key.pem and slicer_crl.pem in the plugin's config directory (or point at them with slicer_cert_pem / slicer_key_pem / slicer_crl_pem in obn.conf). The plugin uses the key to sign MQTT print commands and to install its app certificate on the printer. These are private credentials. This project does not ship them and gives no instructions on obtaining them — you have to find or extract them yourself, and you alone are responsible for ensuring your use complies with the applicable terms and law.

2. Two settings in obn.conf. Set block_cloud = 0 (the default 1 blocks cloud printing outright) and client_name = BambuStudio (the honest default client name is rejected by the cloud print API with HTTP 403).

With that in place you get signed MQTT commands, on-printer app-certificate install, cloud print dispatch, print history and MakerWorld — without touching Developer Mode.

What the default cloud_print = cloud_only actually uploads. Even in this mode the model itself normally stays on your network: for a print with a cloud record the plugin sends the .3mf straight to the printer over LAN FTPS, and only the record goes to Bambu — a project entry, a task entry (mode=lan_file) and a small config .3mf that print history uses for its thumbnails. That record is exactly what buys you the cloud extras: print history in Studio and Handy, and the ability to rate models on MakerWorld. The full model is uploaded to Bambu's servers only when Studio dispatches a pure cloud print (start_print), e.g. for a printer that is not reachable on your LAN.

You can limit even that: set cloud_print = try_lan_first or lan_only to print over the LAN without writing a cloud record at all, and cloud_hide_history = 1 to hide the cloud print history in Studio.

If you run neither Developer Mode nor valid credentials, the printer rejects every print / project_file command and shows on its screen:

MQTT Command verification failed err_code: 84033543

There's been 40 years of documented push back when western governments try to backdoor things - starting with clipper. Pushback is not always successful - but it's far more transparent then China.

It's almost like the real problem is hospital and medical costs versus cost of living.

)which is why as a percentage of GDP, Americans, European costs are remarkably similar, just distributed differently).

from the Summary - ALL monetization across my entire channel has been disabled because of Google's faulty AI and a 'known' glitch within their system. They refuse to offer any way to resolve it. It's also happening to countless others. It could happen to you or your favorite YouTuber. Let's get this out there and pressure Google to fix it.

I didn;t say that it didn't - in fact, the reality was mass casulty events where common in a way that is unthinkable. We do _ in fact_ have plenty of mass graves for colonists. The lack of evidence of the same on the American Indian side makes me skeptical of the whole narrative.

i think it's quite likely that the casualties where in similar scope across two different civilizations, but that the colonialists advantage (if you can call it such) of excess population and higher density of settlers, meant that the colonialists simply outlasted a far smaller population group.

This is commonly repeated, but if you go look at actual record keeping - at the history of early North American and Caribbean/Bahamas colonization through the period where there stopped being as much cross-cultural contact - we see epidemic after epidemic wipe our colonials as much as any. The mortality rate is staggering - even places that should have shown this the most - Haiti for example, the documented evidence was far closer to 100% mortality to Europeans, and significantly less for indigenous. Coming to Haiti during the Haitian revolution was a death sentence for all of the people who lived in squalor in one of the densest cities in Europe - Paris.

Cultures coming together historically were deadly. There is a school of thought that argues that the Persian and Roman downfall was directly associated with the ability of diseases to transit roads.

Of course, this is the stuff that we have records for. It's possible that there was a undocumented mass extinction event, But I tend to be skeptical of historical claims that are based on extremely localized and limited archeological projections.

What do they benefit from taking down any government, NGO or civic work program? American systems, as well as larger European systems are constantly attacked. I've seen the same traffic. Fire walling off China, North Korea and smaller eastern European countries is a must if you ever plan to expose any internet exposed services.

At the risk of noting that this is a PR statement from someone who needs something from AMD - you do realize that you just shat upon the CTO of the largest possible consumer of a technology talking about that technology? It's not 2022 anymore.

Just pointing out, there is a reason the large LLMs keep working to try to make it so their own tools can't attack their own moats, by preventing the exact behavior that you dismiss above.

The 27b dense is _remarkably_ better at coding tasks.

Especially at larger quanitizations (Q4 is pretty crap).

I have a DGX and a Ryzen AI Max 395 - while I love both of them, there are a few critical things that leave the DGX in use, while the Ryzen "just" is my primary homelab server. The biggest thing is prefil numbers, and the performance impact of higher context sizes. Qwen 27b is a great model, nemotron is decent, gemma is workable. But all of them need reasonable context for reasonable outputs.

Unfortunitly, as others have noted, the DGX OS experience... sucks. My hope is that the RTX Spark (which looks to be the exact same stack, sans the high capacity network interface) will help this get a bit more attention, but nVidia's long long long war with the open source community is not helping. Focusing on mainlining kernel support would go a long way to getting the community to be supportive.

Of course, a massive regression just hit Linux 7+/7.1 plus for ROCm hosts, so it's just rough everywhere.

Then this isn’t for you. That’s fine.

The “the Jew was secretly a nazi” this post just spread is one of the most destructive bigoted and destructive anti-Semitic attacks in history

The last three times that Europe had a large independent military force(s), things didn't end so well for the world.

International commerce and business and shared democratic norms are a fscking good things. We have lost the people who learned that the hard way, and they are being replaced with people who have never read a history book.

s/China/Taiwan?

or Tibet?

You are comparing sensational trolls with actual geopolitical history.

The interesting thing is that there is a lot of evidence that Tesla is about to embed Carplay. Rivian has taken over the privileged insanely expensive EV car market, as the Model Y is the closest thing to the old Honda/Toyota dominance, just in the EV space.

This is completely infeasible in the age of mythos. The reality is that the velocity is just not going to feasible from a security PoV without leveraging these tools.

The jet lag team must be in Germany again. Sam, you being deuschbahnned?

Or if you want something from a vendor butting running decade old hardware configs and trying to lock people into their drive ecosystem, UNas or many other options.

Stay away from synology.

This is not a homelab replacement part. It’s enterprise with all the positive and negatives that come with that phrase. The second you start talking about old X hardware, it’s a different product class.

For that use case I recommend UNas from ugreen or the minis forum ryzen Ai stuff.

Synology hardware stopped being decent a while ago.

Most reports I ahve seen have 27b (which is dense) being much more performant then 35b (MoE). Is there a reason for 35b?

The market regulators don't give a fuck if companies screw over their customers privacy wise, as long as it's to the advantage of European companies and European customers.

This can lead to absolute insanity as companies try to satisfy both privacy and market conditions. It's not simple. How many years did google waste with Sandbox?

From personal, separate experience with Europe. It's quite common that the market watchdog and the privacy watchdogs are at odds with each other and make it impossible to achieve a solution that satisfies both of them.

Another example here is Google Chrome - which still allows third party cookies because even thought the privacy regulator wanted them gone, the market regulator required them to architect a solution that was unworkable to not take advantage of their gatekeeper advantages when others didn't have the same rights as them. Google finally said fuck it, and walked away from the privacy features in order to satisfy the anti-trust regulators.

Not shipping this feature in Europe is a common way to deal with satisfying the balkanized regulators there.

Ignore eletrek's self-serving post, go watch the youtube video.

https://www.youtube.com/watch?v=j5oyVNjrUPI

This isn't really true. AI laws in the EU mandate that Apple give full access to everythign on the device to third parties.

It's legit to be skeptical on the privacy front, but giving deepseek access to my entire phone. Or the TrumpAI at some point in a dystopian future seems... not great.