The opposite, if anything. The beam widths used for practical high bandwidth laser communication are extremely narrow. An alien civilization would have to (1) be in the exceedingly tiny fraction of the sky the beam covers, (2) have an unfathomably massive telescope with a correspondingly unfathomably tiny field of view capable of capturing enough photons to make a statistically significant detection, and (3) have that telescope pointed at the exact spot in their sky where the spacecraft is during an interval where it's transmitting. The odds are quite literally astronomical.
In the USA there is The Genetic Information Nondiscrimination Act from 2008 that bars insurance discrimination based on your genes. I assume this will become a landmark law in the next 50 years as society becomes familiarized with routine blood (I mean gene) tests to drive preventive care to the next level
This is also an issue with subways. NYC subway, for example, originally had two different owners and they build their separate networks with different width so that one rail would not use the other. Today, they call it Division A and B trains.
The US South had 5ft gauge during the civil war. There was a massive post-war effort to standardize the gauge that took place over a 36 hour period. All the southern railroads not already on standard gauge did so all at once from May 31-Jun 1 1886.
Tens of thousands of work crews pulled the pins of the west rail and moved it 3in inward. Simultaneously crews at rail yards modified the primary rolling stock.
Service resumed Jun 2 on standard gauge and for the first time passengers and freight could cross the boundary without a changeover stop.
It's a little more complicated in the case of the NYC Subway because the size of the tunnels and stations is based on the size of the rolling stock. BMT/IND (B Division aka the letter trains) trains simply can't be driven through IRT (A Division aka the number trains) tunnels. A "changeover" would entail widening tunnels and platforms many feet below Manhattan, not just moving a lot of rails.
Yeah that makes sense. I just thought it was really neat that the US pulled off such a massive project to standardize gauge over a single weekend - similar to Sweden switching from LHD to RHD on a single day. It's a huge coordination, organization, and training challenge. In the rail case it was done entirely on paper with a little bit of telegraph support.
that's certainly interesting. i just heard about a case in germany where a city has two different systems, and they bought new trains for one of the systems but forgot that they had stations serving both systems where those new trains didn't fit because the other system had narrower trains and wider platforms. the new trains would scrape the platform edge and take damage.
I still do not understand how local rail ends up fragmented like this. The Tokyo metropolitan area has like… what? 8 operators? You’d feel like there would be some consolidation and yet.
In NYC, the MTA runs both so they are technically consolidated, but changing car widths on a live subway system that is used by millions of people every day is on the list of very difficult problems.
For example, if you wanted to conform the IRT lines in NYC (the numbered lines + the 42nd Street Shuttle) to the BMT/IND lines, you would need to (at minimum), update every platform (to back the edge away from the train), probably update every place two tracks are parallel and too close together, probably also update some of the routing and tunnels to enable a larger turning radius, and reapprove all of the tracks (including elevated portions) to support heavier trains.
Since all of the lines (barring the 7 and S) overlap at different points, you can't just make these mutually exclusive changes to one line at a time; you would have to shut down the whole system to make the changes, eliminating service to huge areas of the city for months (if not years).
Like any public utility, there's a huge cost to get the infrastructure in place because there is... well... a CITY in the way. The immense capital outlay is generally worth it for the first to enter the market in a given area, because they stand to collect 100% of the demand for the service they offer. In contrast, a second competing entity incurs the same capital outlay, but can only count on as much of it's competitors business as it can wrest away. Unless their offering is of substantially better value, the most they can realistically hope for is 50%. In cases like local rail service, the initial outlay is so immense that there's little chance that anyone could make a compelling business case for it.
See the earlier comment, the large Japanese private railways are real estate and hospitality companies with a train side business. Each has built their own little fiefdom, but they play nice enough (thru trains, tourist passes, and regional transit payment cars) to make all the effort not worthwhile.
Although during WWII and for some years after all the lines to the south and southwest of Tokyo were under Tokyu Railways (then known as Dai Tokyu, or "Big Tokyu).
They may appreciate that competition between the operators is a mechanism to support efficiency and progress in infrastructure. Each rail operator wants people visiting their malls, building near their stations, and in working towards that, they improve their systems.
If I'm not mistaken, London also has two different types of tube trains, although in that case I don't think it'd because of different owners, but one is simply narrower and therefore cheaper to make tunnels for.
Actually it's both of those, and more complex: multiple private companies building at different times in history, using widely diverse tunnelling techniques and depths to suit different needs.
The London sub-surface and deep-level tubes were made by different technologies. Surface by cut-and-cover and deep-level by tunnelling shield. Both were the first of their kind, and deep-level wasn't possible until electric propulsion was invented. The deep-level tunnels are narrow because of the expense of digging by hand.
Congratulations on the launch team! This is a great need and hope the team can deliver.
My questions are:
1 - Do you have any research on false negatives and false positives for your platform?
2 - How do you build trust in your platform so that users will use your results (and their users will trust it)? Fake news has been so widespread and people continue to believe in it, so why is that any different than with deep fake?
3 - Why are you trying a consumption-type of pricing? Cybersecurity typically charges on a per seat, and it would be very hard for a malware provider charge by 'malware detection'
1- Each model looks for different deepfake signatures. By design, the models do not always agree, which is the goal. We are much more concerned with false negatives, and we target a min of 95% accuracy for our model of detection models.
2 - The challenge is educating users about results without requiring a PhD. Our platform is targeted for use by junior analysts in cyber security or trust and safety.
3 - This is a good suggestion. We are exploring how we can offer an unlimited plan that can cover our high compute costs (we run our multiple models in realtime).
Why would you be more concerned about false negatives? Wouldn't false positives erode trust and value in your product, and considering the applications you're targeting, possibly open you up to lawsuits if you start accusing innocent people of being deepfakes (which, IMO, currently seems unlikely)?
We provide a probabilistic percentage result that is used by a trust and safety team to set limits (ie. flag or block content) so it is not a binary yes/no. We search for specific deepfake signatures and we explain what our results are identifying.
So... that percentage sort of 'return type' allows the people using your service to decide how aggressive they want to be? Smart. It also could possibly turn your service into more of a tool and less of something that someone could blame incorrect results on.
"we achieved the highest quarterly operating margin among all volume OEMs, based on the latest available data, demonstrating that EVs can be more profitable than combustion engine vehicles."
I don't mean this to sound snarky, but I just can't think of a nice way to put it... it seems obvious, based on everything I've ready, they're building these things faster/cheaper and worse than many other cars, regardless of the engine. When you're assembling your cars with zip ties and wood, you're certainly going to be profitable.
So many things I've read about Tesla quality makes me think that they are just not well built vehicles. I know I didn't bother looking at them when I bought a new car last year.
(I have no skin in this game, just a thought from a pretty casual observer)
Tesla makes improvements at a much faster rate than conventional car companies, which results in what we might call 'hacks' as those changes matriculate through the other design components. It's not ideal, but decades of building software has pretty conclusively shown that agility and shipping quickly is a better practice than waiting for perfection.
You might say "well agility is great, except in my self-driving car" but I would remind you that people said the same thing about security software, infrastructure, etc. and were wrong every time. My believe is cars will get safer and better built faster with them innovating than with the traditional model of annual incremental releases. As a Tesla owner, I can also comment that the car feels no less well built than any other car I've owned.
> Tesla makes improvements at a much faster rate than conventional car companies
Is that really the case? I heard their motors are pretty good. But the batteries are mostly Philips and there is a lot of innovation happening thorughout the industry. The software is in parts miles ahead, but their latest UX dicisions are getting dumber and dumber.
> shown that agility and shipping quickly is a better practice than waiting for perfection.
I would argue there is a sweet spot, which Tesla is obviously missing.
> You might say "well agility is great, except in my self-driving car" but I would remind you that people said the same thing about security software, infrastructure, etc. and were wrong every time.
Except for example with the 737 MAX development or the whole "a package got deleted from npm and broke everything". There is a reason for LTS-versions and thorough testing, especially when lives are at stake. My best counterexample to your argument would be Teslas "Full-Self-Driving". It's next year for the last five years, totally insecure and unpredictable, killed people and isn't even better than what the competition has to offer.
If consumers are buying them, it’s somewhat of a moot point. How long did legacy automakers make poor quality vehicles for and still had demand for them?
(own Teslas, all mostly well built, the newer the better)
I think the Tesla has always gotten an unfair amount of negative criticism in my opinion. Yes they have had quality issues, but usually those go down as the production line stabilizes and they usually come out and fix it for you if there’s an issue.
I myself put my Tesla under a microscope trying to find every unsatisfactory item on the delivery day and I did see some uneven panels. But my wife also got a Subaru the same year and I put it under the same kind of scrutiny and it also had similar differences in paneling. Especially under the hood, either the design of parts wasn’t that great or they just put some non standard items to hold things together. It didn’t seem any better than a Tesla (or any cheaper).
I understand how the perusing the media gives this biased feeling, just feels like Tesla mentions make for a more engaged article.
I ordered a MYLR without ever sitting in one due to how neat my friend's Model 3 was. After driving it for the first time, I doubt if I will buy another car. Tesla just makes amazing cars period. Much better than the 3 series I had before as a comparison.
Where does the acronym "MYLR" come from? (Not what it means, but where you got it from.)
I know I can slowly work out "Model Y, Long Range." But it's not intuitive and you kind of have to be a "Tesla insider" to get that far, or at least closely follow their options/trim levels.
It's a bit better than the "M3" I sometimes see thrown around, because an M3 is a BMW, through and through. The car is actually a Model 3, and there's no reason to abbreviate it further unless you're on a Tesla forum (in my opinion.)
Maybe we should have more zip ties and wood in this industry, then? It's frustrating you've formed such a strong opinion without having inspected an actual car, or investigated why people love these things so much.
Zipties are all over the place in any modern car. They are used especially to hold wiring and hoses securely so they don't flop around and abrade against other parts of the car.
Why? If the fastener works, what is the issue? This is the exact sort of thinking that compounds into large delays.
In this case, Tesla obviously ran into an unexpected problem, and did not have a part to fasten this piece. They used various fasteners to secure it, and I'm not aware that anyone has had any problems with it, except that it seems strange.
At the time, Sandy Munro saw something similar in his tear-down, and didn't think it was a big deal. Obviously not a permanent part, but nothing to be worried about, either.
This is the kind of fix you can get with a recall or late problem. It's not a "zip tie" it's a steel cable tie. The wood's a bit shonky but another manufacturer might have got plastic clips made instead of the wood and otherwise made exactly the same fix.
I don't think zip ties and wood are a structural part of the car. It look like it was just there to absorb vibration when they ran out of the foam or plastic part. Not great, but hardly deserving of all the criticism that it is getting.
At the end of 2020, the United States had 1,117,475 MW — or about 1.12 billion kilowatts (kW) — of total utility-scale electricity generating capacity and about 27,724 MW — or nearly 0.03 billion kW — of small-scale solar photovoltaic electricity generating capacity.
Solar went from 0.4 million kW in 2020 to 47.8 million kW in 2020
Pretty impressive S curve, still in the early-stages imo
