Last week, the news was filled with stories of Jack Sweeney and his Twitter-bot that tracks the comings and goings of various billionaires in their private jets. This caught the attention of the billionaire-iest of them all, one Elon Musk, who took exception to the 19-year-old’s feat of data integration, which draws from a number of public databases to infer the location of Elon’s plane. After Jack wisely laughed off Elon’s measly offer of $5,000 to take the bot down, Elon ghosted him — pretty childish behavior for the richest man on the planet, we have to say. But Jack might just have the last laugh, as an Orlando-based private jet chartering company has now offered him a job. Seems like his Twitter-bot and the resulting kerfuffle is a real resume builder, so job-seekers should take note.
Here’s hoping that you have a better retirement plan than NASA. The space agency announced its end-of-life plans for the International Space Station this week, the details of which will just be a run-up to the 2031 de-orbit and crash landing of any remaining debris into the lonely waters of Point Nemo. The agency apparently sees the increasingly political handwriting on the ISS’s aging and sometimes perforated walls, and acknowledges that the next phase of LEO space research will be carried out by a fleet of commercial space stations, none of which is close to existing yet. Politics aside, we’d love to dig into the technical details of the plan, and see exactly what will be salvaged from the station before its fiery demise, if anything. The exact method of de-orbiting too would be interesting — seems like the station would need quite a bit of thrust to put on the brakes, and might need the help of a sacrificial spacecraft.
“You break it, you fix it,” is a philosophy that we Hackaday types are probably more comfortable with than the general public, who tend to leave repairs of broken gear to professionals. But that philosophy seems to be at the core of Google’s new Chromebook repair program for schools, which encourages students to fix the Chromebooks they’re breaking in record numbers these days. Google is providing guidance for schools on setting up complete Chromebook repair facilities, including physical layout of the shop, organization of workflows, and complete repair information for at least a couple of popular brands of the stripped-down laptops. Although the repairs are limited to module-level stuff, like swapping power supplies, we still love the sound of this. Here’s hoping that something like this can trigger an interest in electronics for students that would otherwise never think to open up something as complicated as a laptop.
Back in July, we took note of a disturbing report of an RTL-SDR enthusiast in Crimea who was arrested for treason, apparently based on his interest in tracking flights and otherwise monitoring the radio spectrum. Now, as things appear to be heating up in Ukraine again, our friends at RTL-SDR.com are renewing their warning to radio enthusiasts in the area that there may still be risks. Then as now, we have little interest in the politics of all this, but in light of the previous arrest, we’d say it pays to be careful with how some hobbies are perceived.
And finally, aside from the aforementioned flight-tracking dustup, it’s been a tough week for Elon and Tesla. Not only have 817,000 of the expensive electric vehicles been recalled over something as simple as a wonky seatbelt chime, but another 54,000 cars are also being recalled for a software bug that causes them to ignore stop signs in “Full Self-Driving” mode. We’re not sure if this video of this Tesla hell-ride has anything to do with that bug, but it sure illustrates the point that FSD isn’t really ready for prime time. Then again, as a former Boston resident, we can pretty safely say that what that Tesla was doing isn’t really that much different than the meat-based drivers there.
41 thoughts on “Hackaday Links: February 6, 2022”
Bonus points if the Tesla stop sign bug was caused by training data of real drivers rolling through stop signs.
It wasn’t even a bug. They literally added an option exposed in the UI to roll through stop signs.
This. It was a feature. The feature was called “Rolling Stop”. It was intentionally recognizing stop signs and not really stopping at them.
The only wonder is how this persisted longer than one day.
Hubris, one can only assume. It boggles my mind how that could make it to production for any company, it’s a pretty overt “violate traffic laws” mode. I’m really curious what their lawyers said about it, not that we’ll ever find out…
‘Lotta hate for Tesla by the editors on this site, which I think is strange since if Tesla succeeds it will reduce total carbon footprint by about 30%. The adjective “expensive” in “expensive electric vehicles” seems to betray a level of disdain, unwarranted since EVs have a cheaper TCO than ICE vehicles.
Note that in both cases the recall will be handled by an OTA update, which is much *much* less costly than a typical recall from ICE manufacturers. If GM or Nissan has an update, it’s expensive enough to affect the annual profit. For Tesla… not so much.
The editors definitively state that FSD isn’t ready for prime time and cite an emotional anecdote example, instead of relying on NTSB and other authorities. That’s telling us their opinion, without saying that it’s their opinion, instead of dispassionately reporting facts. Sort of like CNN.
The reality is that FSD doesn’t have to be perfect, it only has to be better than human drivers. Current evidence (driven miles compiled by Tesla) indicates that FSD has 1/9 the accident rate for human drivers. That’s under the ideal conditions that FSD will drive, but it’s hard to believe that the number would be 9x worse when all driving miles are accounted for. FSD would drive drunken drivers around without incident so at the very least we should expect it to be safer overall.
There’s also the fact that mistakes made by FSD will be analyzed and corrected, and dispersed with OTA updates making everyone’s driving safer over time. A human making a fatal mistake doesn’t get a chance to learn from their mistake, but a FSD algorithm would.
The Tesla stop-sign bug (rolling stop) probably *was* caused by real driver training data. It’s not clear to me that a rolling stop should be illegal to begin with, or is any less safe assuming you’re aware of other traffic, but in any event this is an example of the aforementioned upgrade path. The algorithm got a little more compliant with the law, good for them.
Why do you think it will continually get better?
It’s a neural network with limited capacity. It cannot continually get better, and unless you start specializing it by location (like humans do) at some point feeding it national information will only make it worse locally.
Also, the accident rate bit is nonsense. If Autopilot shows a 1/9th accident rate relative to humans, it’s because it’s a biased comparison (humans don’t *let* it drive in dangerous situations). It’s physically not possible to have that low an accident rate. You’ll run into the mechanical/weather/animal floor way above that.
Tesla FSD only has a low accident rate because it will only operate in ideal conditions and mostly on highways or major roads. And even then it’s killed a few people in stupid ways.
The Tesla “stop sign bug” wasn’t a bug, it was an intentional feature, called “rolling stop” which was enabled if the user chose the Average or Assertive option in Full Self-Driving mode. If the intersection was clear, the car would proceed through at 2 mph. The link I’ve provided above is a CleanTechnica article from last October showing images of the UI and descriptive text discussing rolling through stop signs.
More sensibly perhaps say:
Use your RTL-SDR as you like, it is no business of the state (whichever state that happens to be occupying your area at any given time), but make sure to have some good anonymiser layers if you then upload any SDR logs to the internet.
In other news, 50 million IPhones were “recalled” for a security fix.
I thought Hackaday would be above the pathetic Tesla-bashing we see in mainstream media. It’s just a software update, nothing to see here.
Dude, with all due respect:
Gives nothing to support your claim besides “A small percentage of 100mil is still a few mil”
“Over 50 million”
“Just over 936,000 units”
– There was no recall of 50mil iPhones, while exact numbers are unclear there’s talk of “a few percent”, it likely tops out at 5mil or less, lets just say 10% to keep things simple & have plenty of wiggle room around ‘a few’
– According to this very article, 871000 Tesla’s were recalled, that’s 93%…………..?!?!?!?!
– Odd how searching for how many iPhones sold in 2021 gives the number you claimed?
– Both the iPhones and Tesla’s in question went back to the factory to deal with hardware issues, what software update are you on about? are you getting things mixed up?
– Im sorry to be so direct but ehh, do you think a software update can upgrade the subpar ‘self driving’ hardware? can a software update make the plastic stick to the dashboard again? can a software update make the battery last longer without losing speed/mileage? can a software update do ANYTHING about any of Tesla’s big problems????
While i hear pretty much nothing about Tesla on the news here (central Europe) i still think its safe to say there’s no “Tesla-bashing” going on in the US either, people pay top dollar for a fancy car, they should expect the car to deliver on its promises, but since it doesn’t & instead it puts people in danger, news media reports about that? dont think it has anything to do with “bashing”??? its just news media reporting on issues with an American brand?
How about this for an argument, did you defend Samsung when their phones exploded? doubt it right? so then maybe dont defend Tesla’s putting their owners in danger either, because it makes you look like a massive shill?
Do yourself a favor, dont respond, i already know you wont appreciate this response (wich is fine etc, im a terrible person, i know) & there is no need to make yourself look worse by trying to talk your way out of the bs you posted above, my suggestion is to just take this input to hearth and consider it in the future.
1) “If you are going against ‘popular’ opinion, bring actual facts”
2) “There is more to news outlets besides ‘fake news’ and ‘bashing’, they actually report on news, go figure”
Have a wonderful day, may bald eagles serenade you etc.
there’s a relatively recent rule of the internet
“hell hath no scorn like that of a teslaite”
repeat after me:
don’t ever talk bad about tesla. tesla is better than everything and everyone. tesla will save us. tesla has saved all of us. tesla is the answer. don’t ever question tesla. without tesla, you’d have no food, no shipping, no reality. tesla is perfection.
yeah, i’ve never met any group so blindly passionate about any other companies products, a company that so willingly ignores established standards and 100+ years of historical reference into how and why something is made a specific way and insists on reinventing that part and process in it’s entirety, usually to break in new and confounding ways. but progress!
i’m plenty happy with the normal dollar store koolaid, but thanks for asking…
Original commenters point seems to be that an over the air software update should not be termed a recall.
Quote from CNN’s article:
“The faulty seat belt chime impacted 817,143 vehicles, including Model S, X, 3 and Y vehicle from the 2021 and 2022 model year.
Tesla has already begun introducing a software update to fix the issue in production vehicles, according to a document filed with the National Highway Traffic Safety Administration. Other Tesla vehicles will receive an over-the-air software update early this month.”
Key term: “over-the-air software update”
There are legal requirements and steps Tesla must take to ensure as many covered vehicles as possible are repaired. It doesn’t matter if the repair is done over the air or not. Absolutely a RECALL.
“recall” is the technical/regulatory term that does apply here, but its a bit of a misnomer (the focus is on documentation, notices and getting the problem fixed, not on the method suggested by the name).
Well thank you Mr. Helper. https://www.youtube.com/watch?v=VwbFuqdcGoY
Love how the only responses posted are people trying to defend Tesla, makes you wonder if these are even actual responses? or is it just some hired marketing team finding this article and attempting to debunk facts because thats what they’re paid to do??
Anyway, parking that worrying thought for now, ill make some sort of counter argument to those:
Even though i live on a different continent im pretty sure there is no such thing as a rolling stop, you see a stop sign, you stop, even if it seems redundant, the sign is there for a reason after all? The fact that ignoring stop signs is even an option means that the car is breaking road laws????
Besides that lets just say it, sorry for all that bought one and love it, but Tesla’s are terrible cars, wich should come as no surprise, no company “just starting” can make a top quality product, that’s not a ‘stab’ at Tesla either, they’ve done some nice work in these past few years, but lets be real here, the brand was doomed from the start & Elon doesn’t have enough money to fix it. Tesla simply doesn’t have 100+ years experience making cars nor do they have the budget for making something that actually properly keeps track of EVERYTHING around it.
Consider this, Google had self driving cars years ago that had a much higher standard of self-driving, but they downscaled the whole project to only be used in controlled environments, while they already had all the legalization in place for public use & they literally took those cars to all sorts of environments across the world. Makes you wonder why they downscaled? maybe because the real world is inherently complex and you simply cant train the car for every possible scenario?
So if Google came to that realization in a fairly short time-span (like 2-3 years at most?) how come Tesla employees to this day still think they can do better with worse hardware (mostly camera’s instead of super high quality/price lidar) less funds (good luck trying to measure up against Google) and less manpower? (again, good luck trying to measure up against Google) The only answer i can come up with is inflated ego…?? “We work for one of the most overvalued companies in the US so we must be the best”???
Again, no disrespect to anybody that bought a Tesla, i totally get it, they do look cool and you’re (sort of) helping the environment by going electric, but this recent decline should come as no surprise, there have been issue’s and recalls since day 1, and instead of those becoming less common they are now becoming more common as these cars simply are not made to last. Again, something that a lot of people saw coming, but sadly a lot of people didn’t. I hope you can all get at least some of your money back.
They may not have 100+ years making cars, but they have more years in living memory making practical electric cars than anybody else…
Also Google self driving was a rather long time ago when the computing power that could be put on the small platform was considerably lower – its easier to make use of shoddier data when you can throw the computing at it. Also worth pointing out that the nature of relying on any one ‘computer’ vision system is nearly 100% certain to fail in spectacularly stupid ways compared to the human they replace – LIDAR is only good when the objects you are scanning actually scatter some light back to your sensor for instance – so shiny at the wrong angles can become invisible, as can the blacker than black paint type effects (for whatever wavelength light you are using), its also impossible to safely use a high enough power laser to never ever get the projection washed out by that focused beam of sunlight off the highrise etc. And even if you magic up a way to fix those problems you have the problem of how the programmers think to train the model – where a visual or even other spectrum passive camera system is more natural so the initial training assumptions are less likely to miss something important.
Not saying any self driving tech is ready for primetime, just that engineering wise its now very plausable to use camera, and they have some advantages which the earlier google self drive would find much harder. It is also worth pointing out that humans get into accidents all the time because they didn’t keep track of EVERYTHING, and expecting the computer to do so when Careless Air Piano freight company might be flying overhead, or that toddler/small child/pet that you can’t see through the parked car steps out in front of you (which is something an alert human might see coming even though they can’t see the small and foolish creature – by the actions of their ‘responsible’ adult, but expecting anybody to react to that situation in time is asking quite a bit, and the computer with their silicon reflexes might just end up reacting just as well). So self driving as a feature should never be about 100% accident free, can’t ever go wrong before its safe to use, it just has to be sufficiently safe – making it quite likely better than half the humans on the road…
Also all product can end up with recalls even from the biggest longest established names in the brand – like recent PSU and Motherboard failures from previously well regarded brands – its not that something fails but how they deal with it that matters. And as the little Toyota with the pedal sticking floormats did recalling and fixing is the right approach, were the faulty and potentially lethal airbags nobody was willing to really to do so (though admittedly when there are not heaps of other company that can ramp up production of safe design as replacements you are damned if you do take them out and folks die from having the accident or leave them in and risk them being the thing that killed somebody in what would otherwise have been a survivable accident).
On the whole Musk to me comes of with the distinct odor of the snakeoil salesman, but at the same time there is no denying the personality as much as you can dislike it has lead to very impressive engineering teams making damn fine stuff – like currently the cheapest and most ecologically sound space launch vehicles going… Might not be as impressive on some technical aspects here and there compared to the competitors, but they do have a damn fine rocket and are using it very successfully, with high launch cadence.
Yes but (and im not trying to be an ass here) how good are those electric cars when everything else is of lower quality, they are a fire hazard during crashes & it needs batteries replaced after what seems to be ~18months or even less? Plus its not like competitors have been slacking off, all big car companies have good electric cars now, Tesla was the one to really push towards electric cars but by now they’ve lost a lot of their ‘unique value’ because you can just go buy an electric BMW or Porsche or whatever, for less.
I do agree though, in the bigger picture they most definably did make some great strides and progress, and i respect them for it (as well as those ‘early adopters’ for taking the plunge) but overall, yea, you hit the nail on the head with the snakeoil remark, that feeling is what i get from Tesla as a whole these days.
“The fact that ignoring stop signs is even an option means that the car is breaking road laws????”
Yeah, just like people do. The goofy thing about self-driving cars is that people claim they’re safer, but they’re primarily safer for reasons that you could happily enforce electronically on human drivers. There’s no reason to believe that drivers who don’t like driving safely will want a self-driving car that drives safely. Hence the entire reason Tesla had a “rolling stop” feature.
In other words, their entire Autopilot branch has entirely the wrong goals. They’re clearly focusing on customer experience (duh) rather than customer safety.
“There’s no reason to believe that drivers who don’t like driving safely will want a self-driving car that drives safely”
Interesting take on things, never really considered that.
And yea i completely agree, that just means they have completely the wrong goals set for their ‘Autopilot’, the whole point of self driving should be the safety aspect imho, convenience & customer experience is just a bonus
Here in the usa rolling stops are extremely common. Used to be that you would get cited for it , but, now even the police do it. If you take the time to make a full and complete stop the other car will ignore you and rol through on your turn to go. Some of us still make complete stops but I was born in 1972 and learned in drivers ed not to. By the way if you drive in chicago il. usa there is whats called a chicago glimpse. It’s where you just catch a glimpse of red as you drive right through the stop sign. Oh yeah the chicago cops do it to.
Like I did with MIR, I sigh at the news that the ISS is going to intentionally ditched. Once it’s end of life and risks can be taken, I wish that it could be orbited OUT, not in. Attach thrusters and gradually move it higher & higher out of the way to either graveyard orbit, or at least the Van Allens. That way it could be preserved for the future instead of destroyed. I suppose it would just rot (very very slowly) and become a derelict, but it would still exist. And a future or far future camera could visit it the way we visit sunken ocean ships.
Much as the political situation with Russia looks bad right now I can’t actually see the ISS being retired that soon – its been getting a constant stream of upgrades and improvement and is likely to keep getting them as its much cheaper to send up just the new bits and pieces than a whole new space station of similar capabilities..
To me it seems the worse outcome with much likely hood (as I’m hoping WWIII is something folks are wise enough on all sides to steer clear of) is the political divide means the Russians pull out, maybe insist their modules are disconnected to become a separate station, but the ISS remains, and is still the ISS as all the other partner nations are not wanting to start again. In that scenario from what I know of the ISS both the Russian and international sides would need some new parts to make up for essential features lost to the other half, but it still way less costly than starting from scratch, and there is still a desire, even need for the orbital science platform – heck now launch vehicles are starting to get cheaper and better again for the first time in decades maybe they will take your idea and push it out further at some point – even more things to learn when you get further from the protective magnetic field of Earth no doubt.
(Russian-built) Zarya is the core module of the ISS. If Zarya disconnects, all you’ve got left is pieces.
Love this idea tbh, but im kind of wondering if the 2031 plan will even happen, it wouldn’t be the first time for them to go “no wait, lets actually keep our space station up there a bit longer”
Realistically there’s no real reason to decommission it (really? small damage means you’re gonna plummet a decades long worldwide project of literally billions of dollars into the ocean? that makes no sense at all!!) there have been many repairs to many things in orbit by now, so it would make much more sense to just refurbish the whole thing, would be a perfect first step at “building in orbit” too.
“Realistically there’s no real reason to decommission it”
Do you mean “mechanically”?
There’s certainly reason to decommission it. It costs a lot to keep it in LEO, because its orbit’s not aerodynamically stable.
I find it kind of amazing people are treating this as if it’s soon. The writing’s been on the wall for a decade: I’m amazed it’s lasted this long.
Should be noted that it might not be decommissioned the way you’re thinking: it’s certainly possible to disconnect existing modules and use them in other space stations, for instance.
Im not treating it as if its soon, i just ‘dont get’ why it has to be dropped into the ocean, refurbish the thing (like 5-8 years from now) and i think we’re good for another 10-20 years after 2031? if not longer?
You’re right that i didn’t consider them reusing parts of it, but there’s currently no mention of that. Lets hope they do this if they do end up going on with the current plan.
There’s been talk of modules migrating for over a decade. It never happened because they always extended the primary ISS. If the termination becomes more real, those talks will, too.
Oh my God, do you have any idea the fuel that would take? That thing’s massive! Plus, it’d freaking *plow* through other orbits to get there!
It’s an interesting engineering question as to whether it’s possible to get it to a long-term stable orbit quickly. Small boosts to counter drag are entirely different than large scale or it changes.
A controlled deorbit to a specific landing point would probably take the same or more fuel/thrust.
Satellites change orbit all the time without hitting each other, it’s primarily an issue with derelict satellites because they can’t make any adjustments to doge others when they get close.
No, it really won’t. A huge portion of it is helped by atmospheric drag.
The ISS is huge, and it’s not designed to be lifted like that. Creates all sorts of issues. It’s not a monolithic structure: you can’t just put a couple of engines on it and go vroom.
1. The ISS is designed for reboost, no one said it has to be boosted in a single orbit. If you ran a standard reboost every single orbit then it can keep raising it’s orbit slowly, with many small burns.
2. Controlled deorbit to a specific point still requires more than air resistance, as it will break apart into many parts with many different are dynamic shapes. So, it would require a carefully controlled kick to make sure it all deorbits to the same spot.
But, I’m just an idiot on the internet, so, maybe I’m wrong.
Putting some numbers to this: ISS is currently at 400 km altitude, with decay time of less than a year if not boosted. An altitude of 1000 km would have a decay time of hundreds of years.
Currently ISS is boosted by about 0.001 km/s delta-V several times a year ( https://www.heavens-above.com/IssHeight.aspx ). To take it to 1000 km altitude would take delta-V of 0.320 km/s.
If I’m not mistaken in my calculations, a single Falcon heavy flight would be able to give about 0.4 km/s delta-V boost to the 444 ton mass of ISS.
Estimating based on the 320 m/s Dv needed you gave; then with an engine with ISP similar to the ssme (Isp of 452 seconds = 4.43 km/s) it would need a fuel tank with about 1/14th the mass of the ISS to make the burn, so, around 30 tons of fuel. Falcon heavy would almost double that with Leo payload listed as >60 tons.
However, given that usual reboots are only a few m/s, a huge reboot of even 100 m/s (10 tons of fuel) would keep the station up there for many, many more years (I think).
It all comes down to politics and the fact they would have to dispose of it at some point in the future.
Who here expects that NASA will have a replacement ready for the ISS? Nine years from now? I am guessing no way given their Artemis timelines, cost overruns, and delays. Especially if they use Boeing. Say what you will about Mr Musk but he is the only one in the space industry that is actually getting things done right now. In fact we would not even by sending people to the ISS on US spacecraft if it wasn’t for SpaceX.
Regarding ISS: Soviet regularly sent up Soyuz for supplies and/or crews. If they still use it in 10 years, a supply rocket could be attached and be used for braking after everyone has left and all important stuffs were removed.
> After Jack wisely laughed off Elon’s measly offer of $5,000 to take the bot down, Elon ghosted him — pretty childish behavior for the richest man on the planet, we have to say.
This is the same person who called one of the Thai cave rescue divers “pedo boy”… and rightly got sued for it.
Sorry, that is not proper adult behaviour, billionaire or not. It would seem from this latest incident, things have not improved much in this department.
Correction: “pedo guy”, not “pedo boy”… not that the latter is any “better” … the latter would still have resulted in a defamation lawsuit that Elon would have lost.
Not talking to someone who you’ve no need to talk to isn’t ‘ghosting’ them. I’ve never spoken with Jack but I’ve certainly not ghosted him.
How is it sensible for the US, China, Russia, et al, to continue filling the oceans with space junk? Why not equip the ISS, satellites, telescopes and orbiting hardware with ample nuclear explosives to safely obliterate them after relocating far enough away from earth? Meanwhile, more space rocks-predating the Apophis asteroid’s 2029 visit-have actually hit earth, like that “fireball” which came close to hitting North Carolina. https://www.npr.org/2021/09/27/1040987995/fireball-meteor-on-camera-north-carolina It’s bad enough that, according to this article, several thousand meteorites collide with Earth daily. But to date essentially nothing has been done by NASA, Musk or any other public or private entity to deflect asteroids of any size, much less the 60 mile long rocks which will totally annihilate humanity.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)