In other news today...
-
@Gąska said in In other news today...:
There was a case in Poland recently where a train driver was preemptively zapped by a Watchbird.
Of course, the suicidal man was still sliced and diced by the train, because now there was no one to stop it.
5 years or so.
-
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
I'd like to note that freight trains here are not itself quite as long as the ones in the US that seem to go from one side of the horizon to the other. That being said, their stopping distance might still be several kilometers, I don't really know.
And of course you're correct that it's absolutely ridiculous in every aspect. Train drivers are traumatized enough by the people who think this is a good way to commit suicide, punishing them for this is beyond retarded.
-
@topspin said in In other news today...:
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
I'd like to note that freight trains here are not itself quite as long as the ones in the US that seem to go from one side of the horizon to the other.
In Poland you regularly see trains with 40-50 wagons. Not sure if that covers the entire horizon, but it's quite large nevertheless.
-
@Gąska said in In other news today...:
@topspin said in In other news today...:
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
I'd like to note that freight trains here are not itself quite as long as the ones in the US that seem to go from one side of the horizon to the other.
In Poland you regularly see trains with 40-50 wagons. Not sure if that covers the entire horizon, but it's quite large nevertheless.
Those are rookie numbers!
-
@Gąska said in In other news today...:
@topspin said in In other news today...:
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
I'd like to note that freight trains here are not itself quite as long as the ones in the US that seem to go from one side of the horizon to the other.
In Poland you regularly see trains with 40-50 wagons. Not sure if that covers the entire horizon, but it's quite large nevertheless.
In the US it seemed that when stopped for one of these things at a train crossing you might as well step out and quickly get some lunch.
-
@topspin in Poland, railway workers have a habit of closing the crossing 15 minutes before the train arrives. No, this isn't a hyperbole. 15 minutes of waiting, AND THEN the train arrives and you wait for 5 more minutes.
-
@Gąska said in In other news today...:
@topspin in Poland, railway workers have a habit of closing the crossing 15 minutes before the train arrives. No, this isn't a hyperbole. 15 minutes of waiting, AND THEN the train arrives and you wait for 5 more minutes.
Huh all the crossings I have ever seen have been automated
-
The US doesn't have a limit on train length nationally, so you certainly can run into trains over 1 mile long. These are pretty rare as a lot of cities don't allow trains that long blocking junctions.
My office is near a pretty busy section of rail and the longest I have seen through here was an ~3 minute wait. However, I have waited upwards of 30minutes for trains to switch tracks at the switches there.
-
@Dragoon said in In other news today...:
These are pretty rare as a lot of cities don't allow trains that long blocking junctions.
Also, it doesn't matter in some cases, since the fine for over length is less than the profit.
-
@Gąska said in In other news today...:
@topspin in Poland, railway workers have a habit of closing the crossing 15 minutes before the train arrives. No, this isn't a hyperbole. 15 minutes of waiting, AND THEN the train arrives and you wait for 5 more minutes.
They don't do it because they are assholes. Schedule tables are shit, communications they receive are shit, and on top of that, regulations are shit.
So it's worse.
-
One time in Clifton, Arizona, a train servicing the mine in adjacent Morenci decided to park, at afternoon rush hour, across both of the road/rail crossings at each end of the canyon in which the town sits, effectively blocking off an entire section of the town from entering or leaving.
-
@da-Doctah and thus the world was saved. May what walked that day in Clifton not walk again in our time.
-
@Gąska said in In other news today...:
@topspin said in In other news today...:
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
I'd like to note that freight trains here are not itself quite as long as the ones in the US that seem to go from one side of the horizon to the other.
In Poland you regularly see trains with 40-50 wagons. Not sure if that covers the entire horizon, but it's quite large nevertheless.
According to a 2017 Reuters article, the average US freight train length is about 2.4 km (1.5 miles), but some railroads routinely operate some trains 2x to 3x that length. One railroad was averaging just over 2 km. These are in neighborhood of 200-ish cars. I don't know the mass of these trains, but the record is held by a train in Australia at 70 million kg (154 million pounds).
-
@Gąska said in In other news today...:
in Poland, railway workers have a habit of closing the crossing 15 minutes before the train arrives.
This seems very odd to me. Not that the workers would close the crossings early, that's totally believable, but that humans would have to do that at all. The US has over 100k public railroad crossings (gradually declining as little-used routes are abandoned and heavily-used routes are grade-separated) and 80k private crossings. Of the ones that can be closed (some, especially private crossings, rely on people to pay attention and not be idiots; since these are almost always very low-traffic crossings, this is not as big a problem as "not be idiots" might suggest), all (or damn close to it) are fully automatic; an electric circuit detects an approaching train and activates the crossing gates. The idea that 100k+ people would be employed to do this manually seems other-worldly.
-
@HardwareGeek said in In other news today...:
The idea that 100k+ people would be employed to do this manually seems other-worldly.
Only some crossings are manned, of course.
-
@Gribnit said in In other news today...:
@da-Doctah and thus the world was saved. May what walked that day in Clifton not walk again in our time.
Hardly. I reported on the time I got stuck in the oxbow myself, but I was only visiting the town on a business trip. I'm sure it's a regular occurrence there. Like, multi-times weekly.
-
@da-Doctah said in In other news today...:
@Gribnit said in In other news today...:
@da-Doctah and thus the world was saved. May what walked that day in Clifton not walk again in our time.
Hardly. I reported on the time I got stuck in the oxbow myself, but I was only visiting the town on a business trip. I'm sure it's a regular occurrence there. Like, multi-times weekly.
Fair enough. It seemed like good cover for a removal, is all.
-
@Boner said in In other news today...:
Nazi sympathiser spared jail as long as he reads Pride and Prejudice
Ben John, 21, will have to return to court every four months to be tested on his reading by the judge
Now that is some pure luck. It's obvious that he was deceived by the title and expected completely different type of Pride and more of the Prejudice.
-
This post is deleted!
-
@Gribnit said in In other news today...:
@Gąska said in In other news today...:
@topspin in Poland, railway workers have a habit of closing the crossing 15 minutes before the train arrives. No, this isn't a hyperbole. 15 minutes of waiting, AND THEN the train arrives and you wait for 5 more minutes.
Huh all the crossings I have ever seen have been automated
If you don't automate the obnoxiousness, you're doing it wrong!
-
@HardwareGeek said in In other news today...:
According to a 2017 Reuters article, the average US freight train length is about 2.4 km (1.5 miles), but some railroads routinely operate some trains 2x to 3x that length. One railroad was averaging just over 2 km. These are in neighborhood of 200-ish cars.
I once got stuck at a crossing in Canada and because we had nothing better to do and the train was going veeeeeery slowly, we counted cars. I don't remember the exact total but I think it was somewhere in the 150-200 range.
By comparison, in Europe I don't think I've ever seen a train of more than 50 or so cars, and often much less than that.
-
@HardwareGeek said in In other news today...:
The idea that 100k+ people would be employed to do this manually seems other-worldly.
I remember when many crossings in the UK were manually operated. I used to love going to the one just a couple of minutes walk from my grandmother's place and watching all the fussing around with gates and signals. (I wasn't very old, quite a bit under 10 but I'm not sure exactly; I also remember the disappointment when it was automated.) Thinking back on it now, I'm not actually sure which crossing it was; if it was close to her old house or her new apartment. (Same line in either case, but about half a mile apart.)
The UK's been more aggressive about grade-separating, even for footpaths, but there's still quite a few left. I believe that's usually because they're in an urban setting where moving either road or rail is impractical. I suspect there's rules about both line speeds and road speeds when crossings are on the line.
-
@boomzilla said in In other news today...:
Proxies are complicated: RCE vulnerability in a 3 million downloads/week NPM package
Pac-Resolver, a widely used NPM dependency, had a high-severity RCE (Remote Code Execution) vulnerability that could allow network…
The PAC mechanism should DIAF. The browser already had a javascript runtime, so they made the proxy configuration a script, which looked like a good idea, but was a really, really bad one because
- most other applications don't have a javascript runtime, so it does not work with them and
- in those that do there is a huge risk of security vulnerabilities due to insufficient containment.
If it was simple map domain pattern → proxy it would have been both safer and actually usable in non-browsers.
-
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
The mass is mostly irrelevant because friction grows with the normal force, so the same wheel and brake material can provide the same deceleration at any mass (except overheating issues if there is too much mass per brake).
However because the coefficient of friction of steel-on-steel (for train) is so much lower than coefficient for rubber-on-tarmac (for bus), the final conclusion still stands.
-
University of Arizona researchers find link between COVID deaths and snake venom
Scientists from the University of Arizona have discovered an enzyme, similar to one found in rattlesnake venom, that could be driving COVID-19 deaths.
Oh, sure, go ahead and blame the rattlesnakes like you always do.
-
@Bulb said in In other news today...:
@HardwareGeek said in In other news today...:
@Gąska said in In other news today...:
freight train. ... the braking distance was measured with a city bus.
That's ridiculous. There are orders of magnitude difference in mass. The stopping distance of a freight train can be several kilometers.
The mass is mostly irrelevant because friction grows with the normal force, so the same wheel and brake material can provide the same deceleration at any mass (except overheating issues if there is too much mass per brake).
This ignores the physics of tires. For vehicles with tires, weight do affect grip, and thus stopping distance.
However because the coefficient of friction of steel-on-steel (for train) is so much lower than coefficient for rubber-on-tarmac (for bus), the final conclusion still stands.
Tires have additional physics apart from just pure friction, that's why it's called grip instead. So a bus vs a train is an even dumber stopping distance comparison with that in mind.
-
Physics aside, trains have additional mechanisms that road vehicles do not. Track sanding, for one.
And I believe even Soviet M62 were modernized with rheostatic braking (not regenerative - power is dumped into resistor banks).
A bus may have a retarder, but while both mechanisms provide auxiliary braking, it's not comparable.
-
@Carnage The weight-per-wheel does matter, but a heavier vehicle generally gets more wheels so the weight-per-wheel does not differ much between vehicles of the same kind. Which again means the total weight does not matter, because each train car brakes itself.
Either way, rubber-tired road vehicles get much better braking action than trains.
-
@Applied-Mediocrity said in In other news today...:
Physics aside, trains have additional mechanisms that road vehicles do not. Track sanding, for one.
Doesn't track sanding just increase the friction that the train has on the tracks? You're still breaking "normally", just with more friction.
rheostatic braking
The friction between train and tracks still sets the upper limit. If the wheels start slipping, the energy is dissipated as heat between the wheels and the tracks and not through the rheostatic/regenerative system.
This used to be somewhat of a deal with trucks apparently. They had several ways of dissipating the energy from breaking (e.g., disc breaks or via the engine). You'd have to carefully balance these to make sure none of them overheats when e.g. coming down from a high mountain.
-
In other news today… somebody says something wrong on WTDWTF and everyone rushes in to add their own ill-conceived conclusions to the mix
Pretty sure trains of any description aren’t generally grinding the rails like they need more personal injuries this week, so who cares how high the friction could be?
-
Record €225m fine imposed on WhatsApp by Irish regulator for ‘severe’ breaches of privacy law
:quality(70)/cloudfront-eu-central-1.images.arcpublishing.com/irishtimes/2RH3C2XNHWRWYPYQYKS3YNZGTI.jpg)
European regulators directed Helen Dixon to radically increase penalty for Facebook-owned service
One way to raise money I suppose.
-
@Applied-Mediocrity said in In other news today...:
And I believe even Soviet M62 were modernized with rheostatic braking (not regenerative - power is dumped into resistor banks).
Did the braking happen on all cars of the train? Or just on the very front and back? Because if the middle cars, which contain most of the mass, don't do any braking whatsoever, then the equation changes a lot.
-
@acrow The impact is not all that great, true. These days I rarely even see a pusher added. Only two-section puller. Or just one section, when running empty. Dunno about
, of course. Maybe they didn't convert all the railway lines into pocket lining like we did. Anyway...@cvi said in In other news today...:
Doesn't track sanding just increase the friction that the train has on the tracks?
It does. I just want to contribute my own "ill-conceived conclusions" (
) on how incomparable the two things are.My old workplace was a stone's throw away from a 13-track railway interchange. With stupid people playing chicken at least once a month.
-
@Bulb said in In other news today...:
insufficient containment
Very few languages are good at getting that right; they often insist on making everything be linked sets of properties for various things, yet that allows you to get access to all sorts of things even if you can't directly name them.
In short, I really don't trust JS (or Python, or Ruby) to get that sort of thing right. I'm also not 100% sure about Java and C#. (C++ and C are just outright unsafe, and never claimed to be otherwise.)
-
@Applied-Mediocrity said in In other news today...:
My old workplace was a stone's throw away from a 13-track railway interchange. With stupid people playing chicken at least once a month.
The term used by the Swiss railways is "personenunfall", and it's not exactly unheard of to delay train traffic for a bit. The fact that they frequently clean up the resulting mess within the hour in the area around bigger cities and resume traffic means that they're not exactly unprepared for it.
At a guess, that would be one of the more sucky aspects of being a train conductor.
-
@dkf Or, in other words, the main problem is
running untrusted code
in a turing-complete environment.
I'd understand simplified scripting. But executing arbitrary untrusted code just about anywhere is asking for trouble.
-
@dkf said in In other news today...:
@Bulb said in In other news today...:
insufficient containment
Very few languages are good at getting that right; they often insist on making everything be linked sets of properties for various things, yet that allows you to get access to all sorts of things even if you can't directly name them.
In short, I really don't trust JS (or Python, or Ruby) to get that sort of thing right. I'm also not 100% sure about Java and C#. (C++ and C are just outright unsafe, and never claimed to be otherwise.)
As far as I know Python, Ruby and C# never made a credible attempt at creating secure sandbox, and the Java attempt is probably considered failed for all practical purposes (there is some level of isolation, but it's not fully secure). Only JS is somewhat credible when the virtual machine does not have implementation of most system calls at all. Except one can still sometimes get out through bugs.
On the other hand everything, including C and C++, can be mostly secured at process level using namespaces and seccomp. But I don't know how far any Windows equivalent is.
-
@Bulb said in In other news today...:
@dkf said in In other news today...:
@Bulb said in In other news today...:
insufficient containment
Very few languages are good at getting that right; they often insist on making everything be linked sets of properties for various things, yet that allows you to get access to all sorts of things even if you can't directly name them.
In short, I really don't trust JS (or Python, or Ruby) to get that sort of thing right. I'm also not 100% sure about Java and C#. (C++ and C are just outright unsafe, and never claimed to be otherwise.)
As far as I know Python, Ruby and C# never made a credible attempt at creating secure sandbox, and the Java attempt is probably considered failed for all practical purposes (there is some level of isolation, but it's not fully secure). Only JS is somewhat credible when the virtual machine does not have implementation of most system calls at all. Except one can still sometimes get out through bugs.
On the other hand everything, including C and C++, can be mostly secured at process level using namespaces and seccomp. But I don't know how far any Windows equivalent is.
Actually, the whole concept of sandbox is basically obsolete, because it always has access to CPU, some memory and some limited input and output. Which used to be considered "useless" - but then came the Bitcoin and ushered a new era.
The only sandbox that cannot mine *coins is the one that cannot have any output, and such a sandbox can be safely deleted.Although... that's an interesting philosophical question for 21st century. If a bitcoin is mined in completely isolated sandbox, does it make any value?
-
@cvi said in In other news today...:
the resulting mess
Eh, cleanup wasn't happening all that much. No intentionally suicidal people, mostly drunkards who wouldn't scale the pedestrian overpass.
But the thing about emergency brakes in Soviet locomotives is that once it's engaged, the entire thing must come to a complete stop before it can be physically re-engaged. I hear old engineers were clever that way because conductors wouldn't give a fuck (especially if it's some animal). This sometimes includes the horn blaring until the diesel is stopped.
-
@Kamil-Podlesak said in In other news today...:
The only sandbox that cannot mine *coins is the one that cannot have any output, and such a sandbox can be safely deleted.
Not true. It can have output, it just must not be able to communicate the output to the malware writer.
-
@acrow said in In other news today...:
I'd understand simplified scripting. But executing arbitrary untrusted code just about anywhere is asking for trouble.
The real way you make things safe is by being extremely strict in what capabilities you grant to the execution environment. Can't open bad files if you have no access to the filesystem at all. (PAC parsing shouldn't need any exposed OS access at all.)
in a turing-complete environment.
You deal with that by just limiting the amount of execution time and memory available. Screw around and the system pulls the plug on you. Yes, that means it might've just been about to finish, but rather than trying to second-guess that the system just tells you to fuck off when it feels like it and can enforce that.
-
@dkf said in In other news today...:
@acrow said in In other news today...:
I'd understand simplified scripting. But executing arbitrary untrusted code just about anywhere is asking for trouble.
The real way you make things safe is by being extremely strict in what capabilities you grant to the execution environment. Can't open bad files if you have no access to the filesystem at all. (PAC parsing shouldn't need any exposed OS access at all.)
Yeah, that's how it's supposed to work in browsers where the execution environment indeed isn't given filesystem access. But then someone came and implemented running the PAC scripts from an execution environment that does (node.js), and does not support creating properly isolated sub-environments.
-
@Bulb said in In other news today...:
Only JS is somewhat credible when the virtual machine does not have implementation of most system calls at all. Except one can still sometimes get out through bugs.
I don't consider the JS sandboxing to be particularly strong, and that's despite it having had a really large amount of effort put into it. The fundamental problem is that it's still all a big linked ball of mud and identifying all the ways that trouble could be caused is so difficult. Auditing all operations to see if they can be abused (whether on their own or in combination) is a pain, but it's really the only way to do it. It's enormously easier if the language is designed to support doing such an audit, and to have exact visibility of what routes are open between the secured domain (the majority of the app) and the insecure code so they're just exactly those that need to be there and no more at all.
-
@Bulb said in In other news today...:
But then someone came and implemented running the PAC scripts from an execution environment that does (node.js), and does not support creating properly isolated sub-environments.
So… something that's insecure by design is found to be insecure?
-
@cvi said in In other news today...:
At a guess, that would be one of the more sucky aspects of being a train conductor.
One of my brother's friend was a Paris metro driver. They said drivers are in two categories: had someone jump in front of their train, or will have one soon. And yeah, it's common enough that there are whole procedures in place as to how to deal with it, both on the practical level for the traffic handling, hardware checking and repairs... and on the psychological level for the drivers.
-
@kazitor said in In other news today...:
Pretty sure trains of any description aren’t generally grinding the rails
-
@Applied-Mediocrity said in In other news today...:
But the thing about emergency brakes in Soviet locomotives is that once it's engaged, the entire thing must come to a complete stop before it can be physically re-engaged.
In US trains, (AFAIK) there's nothing that requires the train to come to a stop before the brakes can be released, but as a practical matter, on a long train it can take 5+ minutes to build up enough air pressure to release them (there's one compressor in the (each) locomotive pumping air into a tube a mile long), the train is going to be stopped by the time they're released. And if the emergency brakes weren't applied by the engineer (e.g., something came apart in the middle), there has to be an investigation of why the brakes were applied and repair of whatever caused it. And if they were applied by the engineer, there has to be an investigation of whatever incident caused him to apply them.
-
@topspin said in In other news today...:
@Kamil-Podlesak said in In other news today...:
The only sandbox that cannot mine *coins is the one that cannot have any output, and such a sandbox can be safely deleted.
Not true. It can have output, it just must not be able to communicate the output to the malware writer.
Not true. It must not be allowed to communicate back as a node in the decentralized ledger under attack at all, or it could incur a benefit to the writer stochastically.
-
@HardwareGeek The same principle applies to the technical side, but the rest is safety procedure. Which is why you have to get out of the cabin and manually operate the bolt to restore line pressure, with a bludgeoning hammer.
-
