I think it's good to explicitly point out that low-effort participation is discouraged on HN. It might not affect the actual "offender", but it's meaningful to the rest of the readers.
>The fact that Ÿnsect failed doesn’t mean the entire insect farming sector is doomed. Competitor Innovafeed is reportedly holding up better, in part because it started with a smaller production site and is ramping up incrementally.
>For Prof. Haslam, Ÿnsect exemplifies a broader European problem. “Ÿnsect is a case study in Europe’s scaling gap. We fund moonshots. We underfund factories. We celebrate pilots. We abandon industrialization. See Northvolt [a struggling Swedish battery maker], Volocopter [a German air taxi startup], and Lilium [a failed German flying taxi company],” he said.
For the moment ynsect was launched in France it was obvious that it was doomed to fail. Like often here, the only real goal was to suck public funding.
Normally, you would start a small business/factory and scale with your business. Especially growing insect doesn't require a "mega factory".
But here, from the onset, they started from scratch and announced a mega investment to build a giant factory. Obviously getting hundreds of millions or even a billion, most from public funding as we could guess.
Like the huge megalomaniac project that, to french people, is the typical example of too huge to make sense step for a startup, that is expected to be a sink of funds.
Otherwise, there is a good article in english if you want straight to the point article about the history and concret reasons of failure in the following link:
Ain’t no way you want flying taxis in Manhattan. If two collide or one fails, you could kill dozens of people.
Maaaaybe instead of the tunnels and bridges, to increase throughput during rush hours, but even then we’re trying to have fewer vehicles in Manhattan, not more.
Also, I cannot imagine what it would be like to go through an intersection during the winter. You would be hit with a wall of cross-cutting wind tunneling down 50 blocks that no airborne device is going to handle well. Absolute nightmare.
I don't really agree with that. Helicopters can auto-rotate, drones can't. If something goes wrong with a drone system, it is going to crash hard guaranteed and likely doesn't have any meaningful control on the way down.
It's certainly not crazy to imagine that you could cut the costs of a helicopter-like aircraft that was purpose-made for relatively short, relatively low-speed, relatively light load duties.
The energy cost during operations is very relevant, too, which is why you see things like tilt rotor designs with wings/bodies to generate lift.
When Airbus was doing the math on these a few years ago, the pilot cost was also one of the main concerns, so it was "autonomous or bust", and they ended up investing a lot on the autonomous side (not just the aircraft but also urban traffic management, etc).
That's not at all obvious to me. Please explain. Both are approximately car sized and both are likely to hit pedestrians.
In a place with fewer pedestrians I'd buy that airborne vehicles might have a higher chance of hitting a person because they could crash somewhere that a traditional taxi couldn't. But when the place is packed wall to wall with people an arms length away I don't think that applies anymore. At least it doesn't seem self evident to me.
Consider that not all malfunctions of a car leads it to crashing into things. The fuel system, the engine, the transmission and even the steering can completely break down and the car will still came to a stop. They are equipped with redundant brakes, and are always supported by the ground.
At the same time an aircraft is much more precarious. If anything in the fuel, engine, transmission, props, or control surfaces go wrong it will come down and fast. They have much more potential energy than a car (because they are high up). They also typically have much more kinetic energy because they have to move faster to maintain lift if they are fixed winged, or they have to have fast rotating parts if they are rotary winged.
In crowded streets cars obviously go slow, and in any case most traffic accidents don't result in a car starting accelerating uncontrollably until it runs into a building.
Even a 20 m flight height means the taxi will reach 72 km/h before it hits the ground.
I’m not an expert by any means, but one of the major impediments I would imagine to flying taxis carrying people is safety; there’s a _lot_ that has to be done before people board an airplane in terms of checks, paperwork, planning, etc.
The dream of “order a flying taxi on your phone and it takes you wherever you want in five minutes” isn’t really compatible with aviation safety culture (at least at the pilot level in the US). That’s not to say it can’t be done, but you probably need a lot of really good PR people to figure out how to say “we want to remove the safety controls from this so we can make money with it” and have people buy it.
aviation occupies a great deal of my attention, and there is a logic to everything that is done, based on actual provable, repeatable results.
anything involved in high volume passenger aviation has to pass reliability tests that will dry your eyes out just reading through the synopsis, nothing is making it to the PR stage.
I splain little bit, pick some fancy country full of rich people flying around, tell them that the US has just ripped the lid off airspace restrictions (again¹), and is now letting some kind of ubber drone thing loose , and quite litteraly instantly there will be calls for all flights going to the US to turn around as all insurance policys for commercial flights to the US will be null and void.
i don't think mountain rescue is asking for a better vehicle. traditional helicopters work.
flying taxi startups, drone companies, jetpack companies, and all the other fantastical flying startyps keep trying to say they have applications in mountain rescue, but i'm pretty sure that's providing a lot more benefit to the flying taxi startup's pitch deck than it is to any mountain rescue operation.
Traditional helicopters also have the effective lift-weight ratios to tackle the density*altitude of mountain rescue that these "air-taxis" have _zero_ hope to achieve with the the vastly lower power-weight of electrical drive-trains and their lift-inefficient multi-rotor designs.
China calls it the low-altitude economy, and besides human transportation there is a lot that can be done. Personally, I believe that propeller-driven devices are too dangerous and noisy, but there might be innovations coming out of China that Europe can't
Bird flight doesn't scale significantly. You can deliver very small objects via bird, and perhaps build a bird-like drone that does the same. But you can't build a human-carrying bird.
It's very possible, and in fact most likely, that it can't scale. Insect flight is an even better example - the mechanisms that allow most insects to fly simply don't work past a few grams of weight. So, it is simply impossible to create an insect-like drone that can carry a human.
I expect the exact same is true for birds - the kinds of effects that allow birds to fly with so little energy compared to a propeller-based aircraft are almost certainly not scalable, due to the fundamental properties of air as a gas. As far as I know, bird flight is made possible by complex turbulence effects induced by the microscopic structure of their feathers. It's very unlikely this effect could skale to 100kg of weight.
Lower noise, lower operating cost, lower purchase price, easier to pilot, more reliable (fewer parts), safer (redundancy), no emissions, faster time to air, configurable to requirements, etc.
Unfortunately Manhattan doesn't seem like a great place for bikes. The weather is just too variable. Some daredevils will be out in any weather but for most people it's just not feasible about half of the days of the year.
Not that helicopters make any more sense. The city needs some car bans, and yes, bicycles are part of replacing that. But only mass transit will be able to move enough people when there's a foot of slush on the ground.
Weather in New York is perfectly fine for biking. If you can walk outside you can bike. Both means of transportation are equally resilient to bad weather. What you need is protected bike lanes, so you can bike relaxed and holding an umbrella if necessary, as millions of people do every day in Netherlands and other European countries.
Because the need is fulfilled adequately. They are not solving anything new or revolutionising anything old, these are dumb ideas for dumb people to throw money at hoping it sticks.
It is not. Wilderness rescues are extremely resource constrained due to the costs involved coupled with the fact that those in need of rescue were fully aware of the risks before they set out. There's a severe limit to how many tax dollars will go towards bailing adults out of situations of their own making. Lowering costs would quite literally save lives.
At least with respect to aviation, we don't have any non-combustion power-trains that can remotely come close to the power-to-weight ratios of turbine engines.
The earliest cars were replacing the animal muscle power of carriages--a trivially easy feat given that the most primitive steam and combustion engines easily 10x both the raw power, power-to-weight, and power-density of a team of horses.
Agree. It doesn’t have the futuristic vibe but an urban gondola type system is probably what would be best. Especially in a city where there may already be a network of structures to leverage (eg. The buildings/rooftops and elevators). It would require massive coordination or eminent domain type laws to force but end result could be pretty awesome
It's hard to pick just one reason, but off the top of my head:
* Any failure tends to turn flying things into unguided missiles
* Noise is extremely hard to control -- I did an FAA helicopter discovery lesson, and oof
* Cities tend to have difficult to manage wind currents and hit-or-miss visibility. I was in a skyscraper across from one hit by a helicopter trying and failing to land in 2019 -- there's reasons for city no-fly zones
* Limited landing sites makes them highly situational in the first place, unless you want your streets to be helipads, which you don't
These are all fairly intrinsic and not mitigable. I can think of more issues more in the sticks, but you get the idea.
There are about 80,000 non-essential helicopter flights in Manhattan annually -[0]. That means a) there is a lot of demand, and b) it’s been pretty safe, with accidents being very rare.
Many people are against helicopters on the grounds of noise, safety and pollution. Electric taxis will be welcomed once they are certified and economical. They only need to do better than helicopters.
> Electric taxis will be welcomed once they are certified and economical.
Do you believe helicopters are noisy because they're not electric ? Your electric taxi will do the same thing: they need propellers. Propellers that can carry up to 1 ton are fucking loud.
Electric taxis will never be welcomed because they are a dumb idea.
The wind in NYC is no joke. In brooklyn yesterday there were gusts so strong that car alarms were going off. In some apartment buildings, the handicap-accessible automatic doors simply cannot open into the wind.
One more reason is that it cannot actually solve the traffic problem that it claims to solve. It might be able to solve it for rich people when they are the only ones that can afford to travel by air, but if the cost ever comes down low enough for the masses to afford it, I don’t see any reason that congestion wouldn’t be as bad or worse than it is now. And to me it’s not a good investment to improve things just for rich people.
There’s just a lot more space when you can move in three dimensions, so I don’t think the congestion limitations of non-flying cars are likely to be replicated. IIUC (I’m no expert) that’s one of the most attractive features of flying VTOL vehicles.
You're bandwidth-limited on a sparse serialized landing site map no matter what, and you need far higher distance margins that will eat up basically all of the dimensional advantages.
If ground vehicles side-swipe, it's just an insurance claim. If flying vehicles sideswipe, it's a Problem(tm).
I honestly think the most attractive features of VTOL vehicles are that they are from sci fi, and you can look up and see a bunch of empty space and wish you were there while sitting in traffic.
I am (usually) not willing to assume that the founders of highly technical startups would not consider something that I as an outsider would in the first 5 minutes of engaging with the topic.
That makes me skeptical of all of these (minus the wind currents in cities, that might have taken a little longer).
Founders can be chasing a dream and in doing so mesmerize investors. Or they capitalize on that same dream being the investor's. Even if it's not viable, it can still be really fun company to work for and/or earn money at. Even if there is a small lane for that sort of flying machine, the sheer number of companies purportedly working on something like that is suspect. Given the huge costs for development and certification, and the small number of vehicles that will really get deployed (certainly for the first so many years), there must be many that are never going to make their money back. I worked for a drone-adjacent company and now my LinkedIn is swamped with these startups.
Here's my sanity check when reading something like this on hn: What do you have me believe about the founder/investors? I understand that it's fun and common around here to be arrogant enough to presume that other people are absolute idiots, who are incredibly bad at their jobs, but I am not interested. If all you can bring are "duh" ideas, then that's a red flag.
Unless you can bring really insightful ideas, I am going to err on the side of the people who put years of their time into it and the people who put millions of dollars into it
Are they still going to be wrong? Of course. Am I likely to think the sidechair hn commentator is simply missing something in the bigger picture? Yes (and I can think of multiple concrete things in this case)
If a startup were able to truly solve the first two issues alone, they would not be burning those world-changing engineering solutions on flying taxis.
I don't know if a silent, fail-safe, and efficient method of flight is physically impossible or not, but I do know this is low on the list of applications it would be first seen in.
EDIT: I'm looking at the air taxi companies this thread started with, and no, they have not solved any of the relevant problems.
Theranos was famously founded on pitches about blood testing from finger pricks that literally any phlebotomist and many people with a modest life science background could've told you were physically and statistically impossible on their face. You should be considerably less credulous toward startup grifters.
The reason why you (and everyone else) knows about Theranos is that it was unique, which serves as a bad signifier if you want to judge what is likely to happen with the next startup. Being in prison and losing billions of dollars is just not something most people get excited about.
The reason we know about Theranos is because they took the grift up to a huge level and went from grift to outright fraud once they had to show actual results.
It is not only not unique, but in fact extremely common for startups to be grifts around impossible technical promises, live a few years off gullible investors who have way more money than sens and/or for whom losing a few million dollars on a long shot is just as bad as me wasting a few dollars on a gizmo off Temu I know probably won't work, and which then die out because their ideas obviously couldn't work.
They even sometimes find a niche by pivoting to some vaguely related tech. Say, while flying taxis obviously won't work, a startup trying to build them might find itself developing into a small company building helicopter propeller blades for some specific niche.
It’s moronic to have the government pick winners. Only private investors with actual skin in the game will pick those with true potential. This error happens again and again and again
How about funding some housing for the people? Why is it that every city had new huge neighbourhoods built en-masse until the 1990s, and then suddenly stopped (with a few tiny exceptions)?
You've prompted it by giving it the learning sequence from the post you're replying to, which somebody who needs the tutorial wouldn't be able to specify, and it's replied with a bunch of bullets and lists that, as a person with general programming knowledge but almost no experience writing raytracing algorithms (i.e. presumably the target audience here) look like they have zero value to me in learning the subject.
Perplexing how different our perspectives are. I find this super useful for learning, especially since I can continue chatting about any and all of it.
Now compare that to the various books people mentioned at [0]. It isn't even remotely close.
You spoonfed ChatGPT, and it returned a bunch of semi-relevant formulas and code snippets. But a tutorial? Absolutely not. For starters, it never explains what it is doing, or why! It is missing some crucial concepts, and it doesn't even begin to describe how the various parts fit together.
If this counts as "pretty freaking good" already, I am afraid to ask what you think average educational material looks like.
Sure, it's a nice trick that we can now get a LLM to semi-coherently stitch some StackOverflow answers together, but let's not get ahead of ourselves: there's still a lot of improvement to be done before it is on par with human writing.
I consider the "perfect fortgetfulness" of LLMs a great feature, because I can then precisely select what the context is for a given task. Context is additive, so once something's in it, it's doing something: most I could do is try to counteract it, which is like playing jailbreak.
Then again, this might be just me. When there's a task to be done, even without an LLM my thought process is about selecting the relevant parts of my context for solving it. What is relevant? What starting point has the best odds of being good? That translates naturally to tasking an LLM.
Let's say I have a spec I'm working on. It's based off of a requirements document. If I want to think about the spec in isolation (let's say I want to ask the LLM what requirements are actually being fulfilled by the spec), I can just pass the spec, without passing the requirements. Then I'll compare the response against the actual requirements.
At the end of the day, I guess I hate the automagicness of a silent context injection. Like I said, it also negates the perfect forgetfulness of LLMs.
reply