NamTaf

Those steps up already exist. The railway I work for has in-cab radios that we routinely measure the signal coverage to confirm minimal black spots, plus there’s also GSM phones (as in, hardwired in the driver’s cab), in all our locos.

This is hilariously hand-wavey of the complexity and strikes at my frustrations between how most IT engineers see the world vs how real-world engineers do.

There are parallel tracks within metres of one another. Switching yards have tonnes of them and require precise knowledge of what’s in what.

GPS isn’t SIL.

Trains operate in cluttered environments with reflection and echo abound. That causes drift above the separation of tracks.

There’s way more than “a few” dead spots, and what happens if those repeaters fall over?

There’s tonnes of reasons why high tech isn’t used in train safety-critical systems. The fact you suggest an arduino as an appropriate platform for this is frankly absurd and underscores your misunderstanding of the whole issue.

Close enough isn’t good enough. Crashing doesn’t just mean you reboot or reload the program and try again. It means people die.

Edit to add: A good rule of thumb I’ve learnt after 15 years in the game: if you look at the solution to an engineering problem and then start a sentence with “why don’t they just…”, you probably don’t understand the problem well enough.

What system do they use for this? I’m not aware of any remote-stop functionality outside of driverless railways but that’s not to say it doesn’t exist.

SAP developers aren't paid because they're more talented than other developers, they're paid because the niche in which they're deeply embedded extracts extreme business value.

It may take just as much talent to write interfaces for obscure audio hardware and deal with janky bluetooth edge cases, but that doesn't mint the same $$$$ as a SAP guru who refactors a business process to be more efficient.

If you’ve not watched it, go find James Burke’s “Connections”. It takes this idea writ large; technology is almost always standing on the backs of giants before you and changing one thing which pushes the ball a bit further forward.

Sure, but my point is that you had to actively go out and find it. You didn't get an in-lined feed of social media 'suggested posts' that gradually took you from being angry at the machine to some radicalised position.

My broader point is that you can replace that scenario with just about any subgroup with an extremist representation and it holds. Incels are a prime example. QAnon too. The list goes on. The people who end up extreme in those viewpoints don't start there, they get taken on a journey to there though lots of very marginal successive influences.

A social media algorithm is purpose-built to achieve that because it turns out that is the exact same pattern as 'you hovered over a cute cat video once, so we guess you like cats, so we're going tto bombard you with cute cat vids to drive up your engagement'.

The difference is that hyperpowerful social media algorithms have weaponised it. Kids in the 90s had to actively seek out whatever propaganda or whatnot that they wanted to read about. If you were a pissed-off goth kid wanting to read how to bomb something, you had to genuinely work to find shit like that.

These days, you linger on an auto-playing video too long (that is quite possibly intentionally engineered to bait you in) and suddenly the algorithm gets a lock on you and absolutely rams more of that shit down your throat.

It's not even just propaganda either. The exact same cycle is what's at play here with preying on children's weaknesses. Mental health, body image, etc. etc. can all be weaponised the same way. And kids are espeically susceptable to it given the age they tend to start getting on social media is precisely also the age where they feel most socially awkward and trying to find their identity in the world.

The hilarious thing is that that excuse is essentially just security through obscurity. Which is especially rich coming from the company whose vulnerability research team famously set a hard disclosure deadline that was shorter than many others at the time.

I could be wrong, but I'm interpreting it as a sort of analogous system to Word's autosummary feature. That is, You put in a complete text, and then have periodically briefer summaries of it (perhaps in sections, such as paragraphs?). You start with the summarised version, but then as the reader stretches out the text, it dynamically provides more detail in the areas that are stretched out.

e: For those like me who just went looking, apparently autosummary was removed in Word 2010, which given I had 2007 til about 2 years ago and thus didn't realise it was missing :)

Here's a description of it: https://www.officetooltips.com/word_2007/tips/getting_to_the...

To some degree, that doesn't matter. An underlying feature of a competent approach to safety in design is that the design must take maximal ownership of eliminating risk to all people in all scenarios that can be reasonably expected to result from the design.

The moment Telsa set expectations by proclaiming it as autopilot, they took the corresponding responsiblity to make sure it did not generate any scenarios which were unsafe. The moment they implemented features that allowed the attention of drivers to drift more than standard driving, they also took responsibility to make sure that the drifting attention of drivers did not place the system in an unsafe state.

This same issue applies to touch-screen interfaces in modern cars. Drivers could always stare down at their radio when there were tactile knobs and dials, but touch-screen interfaces now expect that because they've eliminated tactile feedback. Telling drivers 'just don't look down' misses the point, because it's the responisbility of the car manufacturer to not create a system where that added safety risk is not controlled appropriately.

Correct. The fact they made a robot that could crush a human hand means they paid no attention to this hazard. Competent execution of Safety in Design concepts would demand limiting the grip force to only what's necessary to reliably move the pieces, which almost certianly wouldn't break bone. If that isn't possible, then it would imply the requirement to find some other way to resolve this hazard in the heirarchy of controls.

Relying on a human is the last option, not the default, when it comes to safety. Human adaptability is not a licence to hand-wave away design responsibility. The most glaring example is Tesla, who is unforgivably guilty of this.

This is bog-standard competent engineering in almost all domains of engineering. It is the table stakes-level expectation of a reasonable approach to safety. I'd literally end up in jail if something went wrong and I had been found to not consider these factors.

Software- and computer-related domains of engineering are a conspicuous outlier when it comes to this philosophy.

The system producing the shadow isn't a 2d lattice, because it also involves the sun. Changing the location of the sun relative to the window will change the pattern of the shadow.

If you wanted to really convince yourself, add multiple windows such that the shadow is affected by the 'swiss cheese' effect of the holes of all the windows lining up relative to the sun.

Aha, thanks for that key point, that does change it a lot. I missed that this was in CA and for whatever reason assumed it was all going through the NZ courts.

I am sure there is some form of cross-border agreement between NZ and the US, but I don't know any details about it.

Does Glassdoor even have a legal presence in NZ? If so, why? If not, why are they even bothered by this?

Could they theoretically not just play hardball, ignore the NZ courts entirely on the basis that they have no legal entity within NZ, and see what happens? Are there cross-border legal agreements that give the NZ courts any teeth? Or would it come down to NZ having to mandate that all ISPs block Glassdoor within NZ?

As opposed to all the sheep in famously temperate Australia? :)

The temperature argument doesn't stack up to me. Have you got a source?

It's not popular, and what lamb is grown there is generally also a very mild flavour as far as lamb goes, being less gamey than lamb found in other countries.

The bigger issue is that toxins concentrate in carnivores in a way that they do not in herbivores.

Why aren't they paid out upon termination? Is that not standard policy in the US? You earnt them, after all. Just as they can't take back your earnt wages, they can't take back your earnt leave.

This, a million times. Tech nerds love cooking up elaborate technical solutions to this problem (I’ve seen it here in several threads over the years) but completely ignore the legal requirements and expectations around establishing next of kin and executing probate.

My recommendation always remains the same: don’t over complicate it and work with the existing societal processes. Society deals with people dying all the time. All major companies, industries, etc. have means for dealing this which have established legal precedent and won’t get anyone in the shit by following them. Let those processes unfold and instead focus on providing your loved ones with the means of having what they need to do so.

I would be surprised if you were legally allowed to go and modify assets before the will is executed.

I know when my partner died that I was not supposed to log in to her accounts and just transfer shit around; banks instead have very well defined processes for working out who is the legally correct person to do that and then empowering them to do so.

Remember, banks and other big companies deal with this all the time. They necessarily must have robust processes for doing it with the existing societal/legal systems of establishing who is the ‘right person’.

This has almost no similarities to stuxnet. A more analogous hypothetical attack to stuxnet would be if they repeatedly cycled spinning rust drive heads in a certain way to cause the motors to fail and corrupt data, all the while faking the SMART data of the drive to not report drive head parking cycles.

Your understanding of either this attack and/or stuxnet is flawed.

I was just wondering whether AusPost offered something similar... it's a shame to hear it's not for residential addresses. Almost all of my mail is deliveries of freight from online purchases, so I make habitual use of the locker box and post office collection services. I find them very useful, but unfortunataely for the plain old mail I still have to remember to check the box once every couple of weeks.

Agreed. Step 6 is where it breaks down, because it's redefined A as being one of the two envelopes to being the halfway point between those two envelopes. Until step 6, it's describing two parallel views of the universe (depending on which envelope you first get - hence steps 4 and 5), and then mashes them together in a way that doesn't work.

Take the practical example of $100 and $200: A is either $100 or $200 depending on which envelope you receive first, so the equation is either 0.5*A + 0.5*2A, or it is 0.5*A + 0.5*(A/2). It can never be 0.5*2A + 0.5*(A/2)

If it detects it’s on the same LAN it’ll route locally, so I find it quite good. The only challenge I’ve had with latency is desync when using it for video but using a desktop mic directly into the PC for audio, as opposed to using the phone for both. However that’s possible to overcome too with delay.

My use case was doing remote language classes. I had multiple programs open on my desktop and had my lesson book on my desk in front of me. I was looking at slides that I had to read, with PDFs open next to the slides shared during the video call. Joining on a phone was prohibitive compared to joining from my desktop.

I’d imagine anything where I need to read slides/notes on a screen would suck if done via a phone. In fact, the only time I use my phone is if my video is off and I am listening passively.

Thanks, I’ll check this out! VDO will route locally over a LAN when it can, which helps keep latency down, but its always great to compare options.

Protip: if your goal is to use your smartphone as a webcam, check out this: https://vdo.ninja

Written by some guy named Steve, it’s an incredible piece of web software that uses WebRTC to stream phone audio and video as an OBS input. OBS then features a virtual webcam capability to take that stream and make it a webcam. I can then also use OBS to do whatever processing I want, e.g. making my webcam also contain a screen share or whatever else.

It’s trivial to then load up multiple instances for multi-angle scenes in OBS, then cut between the two. For example, you could have one ‘face’ camera and one ‘page’ camera showing paper on your desk and make a 2nd scene with the ‘page’ camera as the primary and a small PIP view of your face.

It goes much farther than just being an input for OBS, though. For example, it can create video chatrooms of multiple participants with URL parameter configuration and without touching OBS (indeed that’s now one of its primary use cases).

I use it to stream applications/webpages with my partner when we’re apart so we can watch a movie together by creating a high res vid/stereo audio input with no noise cancelling as the movie, then have her and I connect as lower quality, mono+noise cancelling participants. Each of us receives the video and audio of the movie, but only the audio of each other.

There’s heaps of parameters to control video and audio quality, buffering, etc. - just about anything you need.

I stumbled across it when I was trying to get my iPhone to be a webcam early on in the pandemic. There’s multiple apps for that purpose - many paid - but this was so easy and worked so well that it blew them out of the water from a capability perspective.

I know I sound like a shill but honestly I’m just a huge fanboy. It’s one of those web apps that does a job really bloody well, with heaps of flexibility and extensibility. I’m genuinely impressed with it and all the hard work Steve’s clearly put in.

The docs explain a lot of its capability: https://docs.vdo.ninja/

Flick through the how it works and use cases pages, they’ll explain it far better than me.

Guides that show sown of the advanced capability: https://docs.vdo.ninja/guides

I feel you can do something with floor and trunc to get the remainder, then using round to get the sig figs correct. You can make it abs of it at the start, and then only add a negative at the start of the concat if it's negative.

I rapidly tried and I get something a bit shorter (about half the length), but my rounding ends up with imperfect precision and sometimes I get 10m60s rather than 11m0s due to it. Appropriate rounding at the appropriate points would fix that.

Besides this, it's politically unteniable. Australia even just famously arranged to purchase US nuclear-powered subs - which require enriched fuel - without any capability to produce it or any intention of creating that capability. Their entire heavy maintenance and repower will need to and is planned to occur in the US.

It’s largely understood (I am in a pub halfway across the world and have no specific citation) that the FHBG just bumped the purchase price up by that amount. In Australia, many new buyers’ max purchase prices are dictated by their total borrowing capacity (in that, few limit themselves to less), and many FHBs are competing against each other for a similar cross-section of the market, so grants are easily absorbed by total prices because now FHB B can bid [x] higher and will.

I agree that negative gearing is far more impactful, but I certainly don’t believe my FHB grant helped me enter the market. It just meant I paid that much more to a seller for the same thing.