atoav

It depends? In something like a 800TB Ceph cluster a failure of a single 20TB drive is not felt at all if configured redundantly enough. Then it does not matter how long replacing it takes, only how many bucks each TB costs and/or how much electricity each drive needs.

If the cluster as whole can stomach the drive failures without losing data and is overall fast enough, there is no need to run many small drives where running fewer big drives is cheaper.

Every application will have different trade offs

DAC resolution is irrelevant if your analog electronics produce noise in the lowest N bits. This is why I advice against using this as a primary comparison metric between audio interfaces.

Any audio interface with 24 bits has a digital dynamic range of 144 dB. Bad interfaces will have a noise floor of -80 dBm, the theoretical best value you can reach (at room temperature, for an audio band of 20kHz¹) is -131 dBm, interfaces for more than a 1000 Euros can reach around -127.

And this all assumes ideally set gain levels. Since this is rarely always the case in the real world 32 bit interfaces can still make sense, IF you have analog electronics that can deal with the signals to give you some extra headroom on top. But I'd still rather get a 24 bit interface with stellar noise figures, than a mediocre one with 32.

Aside from the proper input Impedance (High Z), the noise figures are probably the single most important thing when selecting a audio interface for guitar applications, especially if you plan to add a ton of gain in software. Of course drivers, latency etc. also need to work, but I have found that even cheap interfaces do a decent job at that nowadays.

¹: note that with bigger bandwidth that number gets worse, so if you record at 192kHz and you don't have analog filters before that cut the frequency band down to audible range you pay for that (proven many times to be inaudible) bandwidth with worse noise figures

IMO this is a total non-issue. I have never had a digital pedal that had any latency issue. The only latency issues I ever had was when I used a shitty 50 bucks audio interface on an inadequate PC ca 2006.

And I am a very rhythmically educated guitarist/bassist and a DSP programmer, that means I am likely more aware of this than the average person.

Also: you know what else has latency? Sound. 1 millisecond per 33 cm. That means from striking a string on an acoustic guitar till hearing it you also got latency, first it takes a moment for the string to vibrate, then the sound needs to travel to your ears. In big orchestras that latency is a thing great musicians account for.

And every halfway decent musician can do this by just hitting the note slightly early to compensate.

Latency is IMO the biggest issue with vocalists and speakers, because the delayed signal mixes with the original signal you hear in your head, depending on the latency this can either lead to weird comb filtering or irritating doubling. But if your instrument is only heard through the electric chain a lot is acceptable before it starts to get really unusuable. And modern DSPs don't really have latency issues if used correctly.

This opens you to the possible mistake of accidentally passing something of value None that then is mistakenly interpreted as a valid disconnect.

In some cases using None that way is totally fine, but it is a trade off and the cleaner way is to use a separate type

You pay the human because you assume they take responsibility in this being correct. If they literally copy-paste without checking they are no better than AI. In fact they may be worse.

Center-dots as dividers are the new tell tale. And periods at the end of statements (e.g. code comments) that previously would never have had periods.

As someone who programmed a campus identity managment service from scratch, both front-/backend without a trace of LLMs, as it was a while ago:

Yes there is such a thing as an inactive student and it is quite imortant for such a system to differenciate the two. However it is not really important to display it to the user, since that isn't really info someone inactive would see anyways.

And this is part of the problem: LLMs have this tendency to spill the beans (=implementation details) into frontend code. Aa someone who had both hats of the frontend and the backend guy, I still think LLMs aren't producing optimal results here.

Ideally you backend code is elegant, extensible and exactly what is needed to get the job done efficiently and reliably. The frontend however needs to be 100% thought from the perspective of the user (and sometimes from the perspective of a malicious attacker). LLMs still get mixed up with that split.

I did many "interviews" as a casting agent for films, so finding the suitable actors for one or more roles in limited time.

Thank you for highlighting the randomness, because I can't understand the (what must be fake) confidence of some interviewers in their assessments. They think someone who aced the interview will be good in their role at work. How often I had to work with what can only be described as sly, but lazy con-men that "really interviewed well", but couldn't even do the most basic tasks, says otherwise. Unless their role is doing interviews all you get from an interview requires some degree of extrapolation towards the real job. Additionally people may have different days, different characters that make them shine or suck at doing interviews and maybe that specific task was just a mismatch and another one would have made them shine.

Interviewers that are not aware of this have not reflected what it means to do their own job in my opinion.

This is why I also try to make the interviewees shine by giving them the best cards possible. To some degree that mitigates the randomness and it also helps soothe overly anxious or excited candidates. They should get the feeling that I as the interviewer try to solve the posed problems together with them, not that I throw bricks their way and try to misunderstand everything they say. I also try to make the tasks in the interview emulate what needs to be done in the actual job/role. Everything else would be silly.

I have actually gotten great feedback on that way of doing interviews over the years.

400 machines could be anything between 400 and one order, btw.

The best looking diagram library where you can explicitly place elements IMO is schemdraw (python). Ironically it is a library for drawing electronic circuit diagrams with some flowchart functionality on the side.

Examples: https://schemdraw.readthedocs.io/en/stable/gallery/flowchart...

Usage: https://schemdraw.readthedocs.io/en/stable/elements/flow.htm...

Well some are, some are not. Some are simply invited to a nice dinner and given the hint that this might be a contact that gives them access to affluent circles. I have heard it all.

Well I did not say this was a good thing. But if your FAA is the deciding factor and there is something airlines want, the naturally occurring thought would be that airplane manufacturers, airlines and the regulators came together to decide on the future of these regulations, always with the premise that the future should be safer and more comfortable than the present.

And such bodies do exist, e.g. the ARAC (Aviation Rulemaking Advisory Committee) or specially formed ARCs (Aviation Rulemaking Committees). Whether they are capable of producing the desired innovation or if not, why, I can't tell.

When I traveled to Japan I realized that many of their vending machine interfaces are totally in the style of literally any Nintendo game UI I ever played: a bit cute, a bit whimsical, but ultimately very easy to read and fun to use.

But the biggest differentiator in my eyes was sound. Where in Europe most non-broken vending machines are silent apart from the noise they make when the product comes out, in Japan every action, every button, and every step produces a host of sounds, often the machine literally talks to you.

Of course machines with sounds exist here as well, but it is often "just" a layer that has been added to improve accessible with no considerations given to the quality and aesthetics of the sounds. The Japanese machines I interacted with all had their very own consistent character.

The technology is not a neutral thing here. Automation bias¹ exists and is especially problematic in this context.

And if your country happens to have a police force that is historically bad at its job, giving them that power seems like a really stupid idea.

¹: https://en.wikipedia.org/wiki/Automation_bias

That is traditionally thought of as the line between Patriotism and Nationalism. One often used example is that a patriot wants what is good for the country, while a nationalist wants what makes the country look good. Patriots will do things that make their country look bad if it improves the country or helps the people within it, nationalists will avoid doing that and rather wave the flag and try to hide the problem.

However I wouldn't call anything on the American political right patriotic and even their Nationalism appears hollow, as many of their political measures actively hurt the standing and appearance of the Nation. If anything they are just marks of a political grift.

I have been a decade long windows user (and still am), I administer a computer room with MacOS and use Linux. So I know all three pretty well.

Try reinstalling that surface pro with windows 11 without a Microsoft account. Windows is okay, but every update introduces new bullshit, bloatware, accounts you're forced to use, UI-changes. Linux is reliable in the sense that you can chose when and if you want to change things and there is always a way to do it the way you want to do it (something that is the big weakness of MacOS).

Additionally KDE Plasma is the best Desktop UI on all operating systems today. Scripting and automation on Linux is killer, ... The only reason for Windows/MacOS is certain software that doesn't performantly run on Linux (or if Linux scares you).

Huh? I just finished Fallout 4 on Linux, a game that is famously unstable and it worked totally fine out of the box. Install steam, install Fallout, go.

I think this is ill-phrased, but true. It isn't about being responsible for wellbeing, it is about legal liability in case of accidents.

If you (as the airline) did what the rules where that everybody else follows and the FAA mandates and the baby dies you can argue you did everything correctly and the rules are at fault.

If you however go your own special, non-FAA-approved solution, typically the burden of proof that this special solution is at least up to par with existing certified standards is on you, the airline and if a baby dies it is on you alone to defend that solution.

Meaning from the standpoint of an airline there is not a lot to gain, but much to lose if they do things outside established standards.

This may sound idiotic, but it also prevents other "creative" solutions where airlines sacrifice safety willingly to spend less money.

A lot of that has to do with the fact that the people who decide to use it fall into two camps:

1. People who do not have to use it themselves (political decides, bosses, ..), but get lobbied to the max

2. People who have to administer it and know nothing else and endure the lasting pain over the perceived infinite hurdle of changing the system

I gladly can use Linux at work, so I have had a stable, reliable and unchanging system since the past 6 years, but others have to go through major pains for each version of enshittification of their OS update. In my case the OS just got better slowly and incrementally.

No what they are is:

You want to keep a building secure, so you have some structure of access and key managment, who has access to what and who do you let in.

Then you let access and key managment slide, your front-desktop lets in some shady people and their is a hole in your back wall.

The solution: add scaffolding around the facade and empower some security service staff to enter every room through every window and chime in on every front desk decision.

I think talking to another human who knows a thing may still be useful for kids at some point, regardless of what AI can do?

Yet, when I go through the skills and knowledge acquired during my school education there are many bits in there that I am now eternally grateful for, that I did not understand the need for as a kid.

I am not sure where I would have ended up if left to my own devices, e.g. as a kid starting first class I was bad at languages and hated it with a passion, in the last class I had the single best German and English final exams in my school district and language had grown on me.

Yet another thing that GDPR makes harder. That law people think forces websites to display cookie banners is actually a law that requires you to gain freely given informed consent by the person whose personal data you want to process.

Whether you do that on a website, app or in a grocery store is irrelevant. Every person needs to be (1) informed about what data you are recording, (2) what you plan to do with it, (3) who else sees it and then (4) they need to say yes to all of that. Oh, and the No option needs to work equally well and you are not allowed to record any data before they say either yes or no.

That would be the law. In practise even many websites don't fulfill it.

Imagine how well that would make you look as a grocery store.

In the sense that it and its ecosystem is high maintenance. When I call something names I do not do that lightly, but Wordpress is probably the single piece of software that has caused me the most problems over my 20 year IT career. You simply cannot put Wordpress into a customers hands, let it run for 5 years and expect it not to blow up in one way or another.

Just today I spent three hours to manually fix a page where a customers site was defaced after they installed and then uninstalled a translation plugin. I had to write a script that manually check every single instance of translatable text there is on the website.

Yes and Wordpress is a pile of garbage.

> How often have we been in a four-way controlled intersection and the green is on for the empty side. There's no way that's impedance matching

Who claimed it was tho? We were talking about throughput. If all your intersection knows is time, it can't decide based on traffic. Which is why inductive loops, camera based systems and so on exist, but (1) they are not deployed everywhere and (2) if they are deployed there may still be a time based limit for the case of defective sensors.

Our traffic systems are already pretty well optimized where we are willing to spend money on them. Note however that if you 100% optimize any system for perfect throughput, it has zero slack to deal with unexpected fluctuations in traffic patterns. That is as true for traffic as it is for networks. Throughput is important, but so is comfort, resilience, predictability and cost. All different metrics you have to keep in mind¹ and all of them can suffer when you over-optimize for one.

The view from inside the traffic that we are part of may make that seem inefficient, but seen system wide the main problems usually are not traffic lights, but harder to change structural issues.

¹: or rather you don't have to keep them in mind, since there are people who specialize on just that and are pretty good at it

"We lose money on ever sale, but we plan to make it up in volume"

Exactly, a road system as a whole has a certain capacity and friction in the system is used intentionally to maintain order. It is a bit like limiting electrical current using resistors in electrical circuits. Sure you could leave them out, but then you will have problems.

> You want a light to be able to clear the line built up each time

Ideally, yes, practically, no. In reality you also have to consider what comes downstream of a traffic light. It doesn't help at all if one traffic light has perfect throughput only to put everything into gridlock in the next node.

That means in most real traffic systems each intersection lane will have to have less then ideal throughput because you don't want to overwhelm the next section, as this could be even worse for the system overall.

If you have some engineering background you can also think about this as impedance matching. If you have a 4 lane road narrowing to 2 lanes you have a sudden impedance increase, leading to flow issues. If you put an intersection inbetween you now have a way to manage that impedance mismatch with signaling times, but you have to both look at the impedance before and after the intersection.

Preventing a dedicated American enemy to get nukes, an enemy of which the American intelligence agencies said there is currently no sign they work to get nukes and with which you had deals to prevent them from doing so.

Sorry, but all you did was:

1. Showed them and the world that treaties with you are worth nothing

2. Showed them and the world, to gain the respect of the US, you need nukes. Decades of loyalty, treaties or good partner have been shown to be worth nothing, while being an authoritarian dictatorship and having nukes are respected.

3. Showed them and the world that having US bases on your soil is a liability and not a protection.

All in all if you were US president and your task was to destroy the US' standing in the world, there is not much more you could have done.