bradtemp

Of course, in most states there is net metering, so it is foolish to store it; feed it to the grid. Storing in a battery will cost you about 15 cents/kWh and more like 30 cents/kWh with Tesla powerwall, so you need to have that large a difference between the cost of power when you store it and when you take it out. It is unusual to have a battery to store all your unused solar output, that would be very expensive. You don't want to store more than you will actually need, and in particular more than you need from 3pm to 9pm or whenever your local high-peak rate is. So it's up to you to measure how much you need then. (You will have solar from 3pm to 5pm or so, a bit later in high summer.)

If you seek home backup, generally you don't want to pay to run your home at full load during an outage. Surely in an outage you won't charge your car, or run your over or dryer, or your air conditioner on full etc. Consider how much power you need when conserving, unless you are expecting an apocalypse style outage. Get enough battery for that.

Few believed it would be pulled off, except the most devoted of Tesla fans. However, the point is they did promise unsupervised, which is the biggest milestone in this technology, and they didn't get there. It's like promising a moon landing and delivering a New Shepard suborbital flight. Yes, it's a safety driver. A safety driver is, in spite of the vocabulary, not expected to drive. They supervise and intervene.

It varies from country to country. 3 phase can let you send more power with the same copper, but it requires the connector be larger and with more pins, so there are arguments each way -- even in the countries where 3 phase is common. Most EVs don't come with onboard chargers bigger than 7-10kw, which can be readily done single phase, but is somewhat easier with 3 phase. But very little reason to use a 3-phase plug standard in North America. Only a few stations would provide it and almost no homes. Most people never charge at public level 2 stations.

Only the plug in North America is in question here. China has its own different plug too. In NA, 3 phase is uncommon. But for any AC plug, what matters is the rating of the car's internal charger. There are USA cars that take 18kw single phase, with Tesla's smaller connector.

No because when Tesla's system has a problem, as my car does a few times each trip, I grab the wheel. And waymo also has cars with safety drivers but also has run a million miles without one. Run Tesla FSD with no intervention for a million miles and it would have thousands of accidents. Thousands, where it was at fault. Waymo had 2 or 3 minor parking lot style no damage contacts in their million. Not thousands

If you read the press yesterday you would see that the large majority of articles took Tesla's report at face value, so it needed much more explaining than you imagine.

You make me sound like Obiwan Kenobi.

That's not correct, they are no more liable than carmakers are if you are using cruise control.

Unless you can show in court that their system was defective. It is designed to not be able to do the full driving task, to need interventions. It performs as designed.

Forbes pays contributors, who do so by invitation by a Forbes editor who likes their writing. Forbes also allows companies to pay to put in stories, but they are clearly marked as such.

It's amazing what people will write by assumption without knowing.

And yes I believe in the value of the technology. But also in the truth which is why I was very harsh on cruise in the story earlier this week.

The problem is the bus front was occluded and thus its position and vector were estimates of growing uncertainty, not direct readings. It is not understandable that an estimate with high uncertainty took precedent over a fully reliable direct reading from multiple sensors

Strategy is valued by a lot of viewers, myself included. As I pointed out the victory by the mathematician Anna Kiesenhofer in cycling was amazing to see, even though there was no dramatic race, because she used more math and strategy than the others.

But it was still better because it was a race, and you could see how far behind the others were. But a nerd is going to love that race.

It's just not the same when they race with the clock, and of course it takes a long longer. It is the most fair, which is what you want from the standard of making an academic measurement of athletics, and that is valuable. But in terms of spectator excitement per unit time, it's just too far down the list.

My error. I thought I had the premium service as an Xfinity customer, and the ads were just showing up because these were sports recorded live. I definitely would have done that knowing. Well, almost definitely. Streaming DVR is just a shadow of what local disk DVR is. I can seek instantly at distances I configure. I can fast forward and see what's happening. I can do 3x fast forward that's perfectly smooth, and 2x fast forward with pitch adjusted audio -- that's so nice for a lot of sports or for boring sections of certain sports.

So commercial free Peacock would only get so far, and downloading the stream and playing it locally is still better. If you do it right, you can even do standard DVR -- watch the stream while it is downloading, so you are only a short distance behind live, and avoid spoilers.

Nice to see this show up on HN. I wrote an addendum, not in that post...

Olympic rant #2: Fake "team" events. We now see a lot of events promoted as team events where the team members need never have practiced together and don't even meet each other at the venue necessarily. These are team events where just add the scores of various competitors, or "relays" where there is no baton, nobody touches. Sadly to me, this has become a common form for mixed events. I have been wanting to see more mixed events, but I wanted to see mixed events where the men and women actually work together, like pairs skating or mixed doubles tennis or curling.

It's very easy to make a mixed team event in a real team sport. You can make a rule like "3 men and 3 women on the ice."

I understand that in some sports, like swimming, you can't have a baton, but it would be nice if they had to touch or something. In the downhill sports they can't easily physically meet. Mixed team snowboard cross had the man set a time, and that controlled when the woman was let out of the gate (later.) But all this is better than the sports where you just add scores and they aren't really a team at all. These events are just an excuse for nationalism. They will always be won by big countries which have the size and budget to send a group of high level athletes good enough to get the best average score.

While computer camera/laser judged pole vault/high jump would certainly be possible, and much quicker, they would be different in that the visible bar clearly creates a target in the mind of the athlete. It also has the advantage I talked about -- the spectator can immediately see, with no clock, judge or computer to aid them, if they cleared the bar.

As I wrote, not all sports should be for the spectator. But if the purpose of the competition is to be watched, as is the case in the Olympics, that is what to design for.

Reality is all perception systems occasionally wink out, but they also have object permanence algorithms that know they are not actually doing that. Humans are not so different. You see a car coming up behind you, you look away, you presume it didn't vanish. A good visualization would be to show where you predict the object is in a different colour.

While Forbes does have some articles from paying contributors, these are marked. Articles from "Senior Contributors" involve paying the contributor, not the other way around. So not really like Medium or Substack (Though these do provide some remuneration to those who restrict their articles to paying subscribers, I believe.)

It is challenging, but so far, have people refused to buy products/services they can't repair? Have all those products died in the market as a foregone conclusion?

Yes, but the road to the mall changes from time to time, and the parking lot in the mall changes (watch a video of Waymo having to deal with a christmas tree lot set up in the parking lot of one of the malls they take a customer to.) No, they won't sell you the self-driving without the update service because they can't pass the liability on to you unless you're very rich.

These systems are expected to be constantly changing and improving, and making use of constant streams of new data on updates to roads, traffic, and more in order to run.

You say there is no way in hell you would trust self driving enough to close your eyes. My experience is most people come to trust it far too soon, not too late. It's not a problem. People trust something when they see it work a couple of times. Of course the car won't be sold unless it's been shown to do a better job than you do. You still might not trust it -- but most people will, even long before then.

Correct. While 30 miles is the low end of the predicted ranges (that's more for the pure multirotor vehicles that are nimble but not efficient) these are vehicles for that urban trip. Pick you up, take you to your destination, and when they run out of battery, do a quick trip to where they can recharge or swap. The cities that are overwhelmed with traffic have plenty of interest in this. Also useful for police, ambulance even fire (just to get crew and maybe a small pump and hose on scene before the truck gets there.)

Neither humans nor robots can treat everywhere like a crosswalk. We human drivers see pedestrians standing on the edge of the road all the time, often getting ready to cross. If we slowed down every time we saw that, roads would flow poorly. On the other hand, we do slow down when we see people poised to cross at an actual crosswalk, and the law demands we do that.

Right now, live humans are better at reading the body language of pedestrians than machines are, and this is an area of research.

My suggestion is that mapping places where people are likely to cross allows you to dial up your caution in those locations. Not to dial it all the way up to what the law demands in legal crosswalks, but not to have it at the most basic level. Of course, once a ped enters the roadway, you will dial up the caution.

On 45mph roads, peds know to expect cars not to keep them safe. You wait for a long clearing in the traffic and you run if you have to. On low speed roads, we cross even with cars coming, expecting them to see us and react.

Big victory for Uber

http://ideas.4brad.com/uber-and-waymo-settle-lawsuit-giant-v...

The disabling system becomes another attack surface, and unless it is pretty independent, is itself disabled by a sophisticated attack. But scared as we are of external attack, allowing the government to shut off all cars is like letting Mubarak shut off the internet in Egypt. That's a bigger danger than foreign enemies in many countries.

Yes, this is the point of the article

Typo fixed, sorry

Humans only drive about 500K miles in a lifetime, though there are those who drive much more. 2 million miles would be a challenge in any human lifetime, but there would be some professional drivers which attain that.

I (the author of the OP) have an article about accidents at http://robocars.com/accident.html looking into many of these issues.

While many people do buy a car every 3 to 5 years, they sell that car to somebody else, and the average car lasts about 19 years, with several owners. On the other hand, cell phones are sometimes resold but not nearly in the same way. Many of us have those old phones in the drawer, and would still do that even if there were a big ecosystem for selling them.

Battery swap failed after a $700M investment from Better Place for many reasons. But the big issue is that with battery swap, car owners do not buy or own their battery. But most of the innovation is in the battery. For innovation, you need competitive markets with lots of buyers and lots of competitors. It's much harder to have that with swap