The expensive part of forest fires is paying back homeowners who lost their homes in places guaranteed to be lit on fire, at prices for homes as though the fires didn't exist. The way we chose to do this is by saying it was PG&E's fault, and in exchange, PG&E gets to recoup those payments via permanently higher rates.
It is a little complicated, but it isn't that complicated. The simple question is, should the government pay a safe home's price for a burnt down home?
No. Let owners exercise owner's responsibility (e.g. insurance, and if insurance is too expensive -- well, the risk is too high).
PS: I heard the thing California does, however, is putting a cap on insurance premiums, so insurers just avoid some regions, and owners cannot find insurance to buy. It's kinda the same thing -- owner's responsibility.
California FAIR is the insurance of last resort so what you’re saying isn’t totally accurate.
There has to be an insurance option because you can’t get a mortgage without insurance. And owner occupied real estate prices do not go up without mortgages.
California bends over backwards to make owner occupied real estate risk free.
More provocative questions: what is the difference between someone who lost a home in a place guaranteed for the home to eventually burn down, and someone who doesn’t own a home at all? In that moment: nothing, right? Why is sunk cost a fallacy all the time, except that time?
Is someone who pays less in taxes deserving of less, more or equal government assistance? No, right? Now replace taxes with “compulsory payments” like home insurance: does your answer change?
This should illuminate for you why CA wildfire bailout policy is so inequitable. These communities are not an escape valve from overpriced real estate in California cities, they ARE the overpriced real estate all the same.
AFAIK in CA insurance rates must be set based on historical trends not anticipated future losses or reinsurance prices. It is easy to imagine why - insurance companies love to play financial games when they can. Historical data is lagging by nature and the reinsurance market predicts large fires will continue thus the insurers get hit from both sides.
Note that a lot of the property insurance regulation stems from a 1988 voter proposition. I suppose it has worked fine from then until now but the CA drought and greatly increased fire risk was an unexpected shock.
FWIW I would guess that we won't see extreme fire events for some time going forward - probably not until a "big drought" comes back to CA 30-40 years from now. The reinsurance market will settle down and mutual insurance companies will end up issuing refunds eventually.
The camp fire was caused by a failed hook on lines where similar hooks showed extreme wear-and-tear, despite PG&E claiming to have inspected them recently. It's not like we just decided to say it was PG&E's fault; their inspections were clearly missing important deferred maintenance.
If the fire had been caused by someone without the funds to pay for damages (e.g. a homeless encampment (Day Fire) or college students improperly extinguishing an illegal bonfire (Tea fire)), then there might be criminal charges, but insurance companies will be on the hook.
So some random person lost their job because they didn’t actually do the inspection and now everybody in northern CA pays higher rates. Do you see what you did there? Who do you think won here?
People will not start doing proper inspections until you punish the individual harshly, instead of the company.
Bad philosophy. Less prescribed burns mean more uncontrollable wildfires which means in the long term costs are even higher.
Prescribed burns are expensive now because we haven't done them for so long. California banned the indigenous practice of cultural burns before it was even a state! But the more we work on restoring this practice the cheaper it'll be for everyone in the long term
Enshittification strikes again. In this case, fees and costs go down by virtue of being pushed out into the future as even higher costs as a result of lack of fees being paid now. Someone should make an encyclopedia or reference doc detailing all the different and specific ways Enshittification manifests. Bonus points if they tie it into Socialism/Communism because I'd bet there is a high degree of overlap between the two in terms of failure modes.
amazing mental gymnastics, describing how western-markets-failure-mode is directly tied to fictional-enemy-politics . More seriously, maybe systems on a large scale are susceptible? we see evidence of this here?
I slightly grow tired of saying this, but also not. It’s not capitalism, it’s focusing resources on the wrong projects. Homeless people have drugs, but half the state is on fire.
As described, it is a fair ways away from what RealPage is doing. Specifically:
* RealPage sells raising rents, not just market info.
* RealPage pressures clients into taking their higher rents.
* RealPage also pressure clients to refuse to rent at lower rates for their own narrow economic interest - in other words, they actively seek to circumvent competitive pressure to keep rents high. (edit: to clarify, I mean they discourage lowering rent to attract a renter)
Pave does sound like it gives businesses a leg up over employees in wage negotiations, but until it e.g. starts promising clients that they will be able to pay lower salaries, the critical element of coordination won't be in the mix. Pave gives you the data, but you can still choose to pay above market to attract talent.
It's almost certainly for the companies to pay less money, but with a more generous reading, I think it could be argued that that doesn't necessarily have to come out of employee salaries. That data could be used to:
- Set reasonable ranges to find the right candidates they are looking for faster and minimize hiring friction
- Standardize payment levels in a way that reduces legal liability in certain states like Colorado/California. Or the most generous reading of "reduces legal liability" would be "promoting fairness".
- Reduce the time spent by HR/other teams of negotiating or setting salaries, as they can simply target some target like "we want to pay more than 60% of companies like us"
- For budgeting/forecasting with new hires, this allows companies to have more confidence in their estimates as they plan hiring.
I routinely get emails like "we'd like to hire you as our CTO, and because we just got a bunch of VC money, we're prepared to offer you a generous comp package of up to $90,000 salary plus .05% equity! Must be onsite in San Francisco."
If they were aware of market rates, they could avoid making potential candidates laugh at them.
Literally the worst experience of my life. I don’t even mention it on my resume or LinkedIn or anything. I suppose it was a success and the product is still doing fine. I just never want to do that again unless it’s for myself.
No, this was a sort of AI assistant before AI thing. It was a whole bunch of state machines and decision trees and NLP. It actually worked fairly well.
Which would be perfectly fine if you made this completely transparent and made the same information available to your applicants. Them not knowing the market rates (at least not even remotely as accurately) puts them at a significant disadvantage and you can't expect that most company won't exploit because it would be irrational to do that.
It makes me wonder how it would affect salaries if companies were required to make the salary distribution public for all their roles. So you knew both the range where you were applying, as well as at other companies.
And also how that would interact with unionization. E.g. would it make collective wage bargaining less necessary to some degree? If workers felt they had the data to know they could bargain individually for more money?
What I've heard from leadership at more than one company is that they choose a percentile of the market they want to pay and then set the compensation there. For example, they may say "we want to pay at the 75th percentile for SWEs with X experience in the Bay Area".
I certainly don't trust that this doesn't hold wages down overall, particularly in the boom hiring market we had until recently.
And how do they arrive at the budgeted number? Lots of companies want to ensure they are paying a sufficiently high number to get sufficiently capable employees in a competitive market. While many (including me) find things like Pave gross, it's not a one way street, they can push wages up.
You’re thinking of the actual budget for a position not what a company could in theory budget.
A small businesses owner who pays themselves whatever is left over after expenses doesn’t care about what other companies pay, the company only has so much money. Apple could increase salaries up to the point where they make zero profit, but the goal is profit maximization not salary maximization.
It’s fundamentally the attempt to limit salaries that causes companies to look at the overall market.
Small business owners aren't the target market and are likely to not use such a product.
Hiring well is hard - it's not super obvious if you aren't paying enough or your company isn't desirable or what else is the cause of not seeing good candidates. While in theory you could solve that by wildly overpaying, in practice you have to be able to justify your decision to higher ups in most cases, and pointing to a tool that shows what you really need to pay to get good people can be very helpful. I still find it gross, but, there are practical situations where it will drive salaries higher.
What you just described is speeding up a process not increasing wages.
If a company simply isn’t offering enough money they aren’t going to get the workers they want. Which then forces the company to either go without or increase their compensation, just like how every other market works.
> your company isn't desirable or what else is the cause
The clearest example of this principle is companies eventually learn they need to pay the asshole tax. Market research may suggest X is a reasonable number, but they simply don’t get enough people without paying appropriately.
> will drive salaries higher
It will drive some offers higher and get workers sooner, but an offer not accepted is a salary that doesn’t exist.
> If a company simply isn’t offering enough money they aren’t going to get the workers they want.
I've never seen a worker walk away once an offer is made. Getting workers in the door in the first place is hard, though.
Which means that it is really the marketing that needs to be improved. But how...
> how every other market works.
Every other market that has suffered from this kind of marketing problem starts to publish sale data! The labour market has just been slow to catch up with everyone else here. No doubt because, historically, getting people in the door was easy. That has only somewhat recently started to change.
Or only the most desperate stumble through my door?
But either way, it's pretty clear that the prevailing problem is finding people at all. Not just me, but most businesses have been struggling with the same problem. You can't tempt someone with more money if you have no way of communicating with them.
Which goes back to the marketing problem. Absolutely there is a marketing problem, but publishing price data is exactly how other markets have solved the problem. We shouldn't be shocked or disappointed that labour is going the same way. That's how markets deal with the problem.
Now that is interesting. I personally find the difference almost without meaning, BUT I am very curious. What is the incentive for them to do this? To they get a point off of the increase?
> I'm confused about why we're comparing GOTO control flow with channels, since they're completely unrelated. I guess you can make a mess with each when you use them inappropriately? With GOTO, you should avoid it because there are better options (notably, conditional statements).
Notes On Structured Concurrency goes into some depth on this comparison:
> Notes On Structured Concurrency goes into some depth on this comparison:
Yeah, this is why we use `sync.WaitGroup`s and similar abstractions in Go. Using bare channels (like using `goto`) is fine for small local things, but if you're building abstractions you should use patterns like those discussed in the article (for example: https://go.dev/play/p/UoB8ECDbaTw --I'm sure this can be abstracted further).
I have a good friend who's a Lyft driver. According to him, all drivers are rated on cleanliness by passengers; if you're dinged for a weird smell, there are lasting financial consequences (even if it was for reasons outside of your control, e.g. using a Lyft provided rental while repairing from a traffic accident).
We'll see how Waymo handles it! It will definitely be Waymo's problem to solve, though.
> Many of our riders choose Waymo for the clean and consistent vehicle we offer. To ensure every rider gets this experience, we’ll be applying a vehicle cleaning fee for riders who leave a mess behind in the vehicle, such as vomit, excessive trash, and smoking odors.
> For those that self-report their mess during their ride (not including smoking), the fee will be $50. For issues that go unreported, we’ll charge riders $100 for the first violation and increase the fee for subsequent violations. Repeat trash and smoking related violations may also impact your account standing.
That’s an awesome policy. Compare to car share services (in SF.. apples to apples) such as Gig Car and Getaround which allow unsupervised general access (i.e. no driver there to witness car treatment)… those are generally pigsties. Blunt ash all over the dashboard, used kleenexes in the door handles and cupholders, trash on the floor. It always blew my mind that the perpetrators weren’t fined into the dirt. Good for Waymo.
Considering that I’m in a social bubble of considerate people all of whom wouldn’t leave a single bottle cap in a car, this makes me despair at how people must be outside of my bubble.
I'd say the vast majority of people don't litter or trash things, but the few that do ruin things for everyone. It only takes a few bad apples (on in this case trashy people) to ruin something for everyone else.
Its like living with 4 roommates, but one of them leaves their shit everywhere. It could make the whole place look unkept for everyone else.
And because we generally don't want to pick up after other peoples shit/mess things may be left trashy for a long time before anything gets cleaned.
I haven't used Lyft in awhile but as an Uber rider what I can tell you is, if that company has a similar policy, it hasn't worked for me. The quality and cleanliness of the vehicle, and the quality of the driving, has been vastly inferior for me in San Francisco than Waymo has been.
I travel a lot for work so use Lyft a lot and I can say the same. Its not every time, but maybe 1 in 10 times I'll get in a car that smells like absolute shit, smoke odor, unclean, etc.
I think theres a stigma against reporting drivers so most people don't, but then that leaves the people who don't care about these policies to continue giving people bad experiences.
I feel that. With about that same frequency, I encounter a vehicle or a driver or driving that's just dreadful. But, I don't report it or even ding the driver because I feel like that's a hard job that I don't wanna do, and I feel bad. I dunno, maybe I should.
I think the difference is that as an individual driver you have an incentive to keep your car clean, so that Lyft continues to dispatch riders to you. For Waymo the selective pressure is less direct and also spread across their entire fleet. They can accept a level of dirtiness, given some probability that the rider would reject the car x cost of rider requesting a (partial) refund x etc. etc.
More cynically, there are simply too many people that won't take care of "public" property. If every 3rd rider (exaggerated for rhetorical purposes) trashes the car, it's gonna be dirty no matter what.
That doesn't cover all the normal cleaning needed just from people being in the cars, on top of environmental stuff (sand, dirt, mud, leaf debris, pollen...)
Like others said: the second they have run drivers out, the cars will stop getting cleaned obsessively because you won't have a choice.
Same thing that happened when Lyft and Uber when they ran the taxi industry out of business.
Same thing that happened when Zipcar was established. Cars went from being spotless and well maintained to damaged, dirty, and half the dash being lit up.
> I used to work at FB and they have a team that tries to catch employees selling access like this.
For folks who aren't familiar with FB, maxrmk is absolutely right. But some more color would probably help:
When one of the privacy teams discovers a violation of this kind, the employee is generally called into a meeting with HR and fired the very next day.
A friend of mine did this inadvertently - just trying to help a real personal friend with an account issue, and inadvertently accessed a system in a way he didn't realized was a privacy violation. Months later, he was investigating data for a project, which triggered an audit. They walked him out the door the next day after finding it.
> and inadvertently accessed a system in a way he didn't realized was a privacy violation
Sounds like they need better controls, there shouldn't be ways to inadvertently access personal data and violate someone's privacy. Particularly not at such a mature company.
I don't work there but I imagine when this happens it's because the employee needs access to the resource for some legit reasons, but accessing it for illegitimate reason is what amounts to the violation. So access controls here would amount to reviewing the reasons for the access.
Solution would then be to ask for and log the reason for the access. Possibly with an approval needed by a second person. You can still lie about why you need access, but at least it is logged then.
The controls have gotten better / more explicit over time. They flash you up a pretty explicit clickthrough wall now. And there's pretty explicit training that you hand off issues for friends/family to a 3rd party engineer to handle rather than accessing user/friend data yourself.
When I worked at Google it was literally impossible to access personal data like this in most roles, even for my own account. So it seems like meta leaves something to be desired if it's a click through and a policy.
But circular references don't leak in Java. You have to have a GC root (e.g. a static, or something in your runtime) somewhere pointing at the thing to actually leak it.
There is one case where a "circular" reference can appear to cause a leak that I know of: WeakHashMap. But that's because the keys, which are indeed cleaned up at some point once the associated value is GC'd, are themselves strongly retained references.
They're designed so that, if necessary, they can be operated by a single employee. That's why they're so tightly cramped around the grill: so that someone can cook and still keep an eye on a table.
This is also what makes it really easy to get attached to your local Waffle House: the staff make the place what it is.
Chet is going to be missed a lot in our community. His talks with Romain Guy were inspirational to me: whenever I saw them it was clear that they could each hold the stage on their own, but that having the both of them got more out of both of them, both technically and in entertainment value.
I hope he does well in comedy, but I will miss his example: that you can bring even unusual talents and interests into work life and make something worthwhile of it
It's a low traffic 4 way stop sign intersection. Low speed, probably a 25mph speed limit?
If you're cycling, this kind of intersection is only dangerous if you fail to yield/stop appropriately to auto traffic. Which seems like might be what happened here, from the description.
Maybe. Article says bicyclist was obscured by a truck.
17th street in Potrero is relatively quiet but that particular area often has cars coming off the freeway. It can be a maze to get through on bike. I wouldn't rush to make a judgment until we had a dashcam video.
Does 4 stop sign mean cars have priority over cyclists? Sounds weird reasoning. But not living in this country I don't understand who would have priority if everybody has a stop sign.
Is it just a priority based on power? 18 wheeler > Buses > Cars > motorbikes > cyclists ?
it isn't priority by vehicle type, but by time that you arrive at the stop sign. everyone has to follow the rules equally. approach the sign, come to a complete stop, whoever was there first goes. if the intersection is full, then opposite pairs alternate with whoever is going straight getting right of way, followed by whoever's turning.
Over here I'm pretty sure that's the only rule. Having whoever came there first go sounds like it could get very confusing in even vaguely ambiguous cases.
In my observations in my state, the "who gets there first goes first" is very much not confusing. What seems to confuse people is that with ties, people forget if the rule is the person to the right goes first, or to the left. So in practice, this tends to be negotiated using hand signals.
For any non-Americans wondering what happens if 4 vehicles arrive at the same time all going straight: Yes, it does happen, and it results in 4 people waving each other through, start-stopping in anticipation of each other, and finally someone blowing through the intersection at full throttle in frustration.
The alternating pairs is just convention and isn't actually law.
The actual law is that you need to complete stop first, yield to any vehicles already in the intersection, yield to the first car (tie goes to your right).
Which allows for an interesting scenario to occur.
1) East-bound truck arrives at intersection and stops first before proceeding into intersection.
2) South-bound bicyclist arrives at intersection and proceeds (illegally) without stopping.
3) West-bound Waymo arrives at intersection and stops but proceeds (now also illegally; as the cyclist is within the intersection) through the intersection.
Alternating pairs emerges directly from those rules. Once one person goes north, a non-conflicting person can go south. Then, the current east and west people have arrived before the new north and south people.
Unless the North & South people arrive at the intersection at the exact same time they must still come to a complete stop. While say the South person is stopping an East person could have finished their stop and since the North car has passed by them they can enter the intersection.
Sure as the intersection backs up it becomes very likely that the North-bound person will be blocking an East person from entering the intersection such that the South person can enter. But just because a car is coming at you in an intersection doesn't mean you can legally enter it.
By convention though, literally every East bound person is going to expect the South bound person to travel at the same time as the North bound one. My point is strictly that the law doesn't require this and under some circumstances the South bound person may be breaking the law.
This is not true in Idaho. A stop sign is a yield for bikes. They call this the "potato" laws.
The practical reality of this is that if you're stopped at a stop sign, a bike may be coming from behind and will pass you on the right as you're trying to make a right turn.
AIUI, in this situation the bike has the right of way.
It definitely hiccups in high traffic times, but in practice it works reasonably well around the country. Its generally not meant for high volume intersections, where you'd expect a light instead. For perspective, I'd imagined roundabout and other designs as better, but hadn't really considered an alternative algorithm for people to follow.
1. no markings at all, for very low-traffic places. You're supposed to be driving so slowly that you figure it out. (Yield to the right, if you can't figure it out. In practise, yield to bigger vehicles, pedestrians, cyclists etc — the roads is so small there might not be much choice.)
2. one road is the main road, the other is minor — minor traffic yields. This is most common. There might be signs, but there might only be road markings involving some sort of triangles (i.e. a yield sign).
3. a painted or similar shallow roundabout, showing roundabout rules apply (yield to the left) but allowing large vehicles like buses to drive over the paint. Only used in some countries.
4. an actual (concrete, kerbstone) roundabout, if there's space for that and traffic is higher.
Europe has an algorithm that, on such intersections, you have to yield to a car on your right-hand side. This works well until there's a car on every leg of the intersection, which is extremely rare (because, on intersections with such levels of traffic, there are road signs explicitly saying which road has to yield to which one).
- first to the intersection has the right of way.
- if you both arrive at the same time, the one to the right has the right of way.
- If you both arrive at the same time facing each other you can both proceed driving straight or turn right.
- If you both arrive at the same time facing each other anyone turning left must defer to the other.
> If you're cycling, this kind of intersection is only dangerous if you fail to yield/stop appropriately to auto traffic.
Or if auto traffic fails to yield/stop appropriately to you.
Of course it's almost always the case that dangerously-designed roads only cause accidents when someone does something wrong. A one-track road with blind bends and no speed limit is perfectly safe if people use it safely.
Some will come and tell you that cyclists have an innate right to ignore all traffic laws and that if you were in Europe the car will always be at fault.
Those pesky cyclists might even argue that traffic laws only exist in the first place because car drivers just would not stop maiming and killing people. Crazy cyclists.
They do say that. Maybe the legal liability is on the driver, but that doesn't mean much. All behaviors together from everyone involved is what leads to accidents. If the cyclist feels that the law is all the protection they need to avoid being hit and enter an intersection without looking, or worse, assuming that someone will stop for them, then obviously the premise of their safety was flawed and they made a decision that lead to their death or injury. The accident happens no matter who gets the blame.
Of course they are different things. But the choice to define "range" as the highest possible number is misleading: people ask the range so that they can know how many stops are required at a highway speed trip.
Using surface street numbers for that is....unhelpful, at best. It certainly doesn't help the Windows computer hoi polloi feel as if they have gotten what they paid for.
I think they should just delete that number and just show energy and energy remaining at destination. It will annoy some luddites but it's the only reasonable way to display the actually important information, i.e will I make it and how much juice will I have when I get there.
Yup. There's a preference to show percentage rather than theoretical range. The latter is much more useful IMO, but percentage should probably be the default because theoretical range is misleading until you understand it.
There’s really 3 distinct numbers that are all useful today.
“Charge” = total energy which is useful for charging but not specific route planning.
“Range” = charge * temperature which is a useful benchmark for known trips and drivers. The specific number isn’t as important vs knowing you can do regular trips between known locations in X units. ie you can commute 4 more days without charging.
“Remaining charge at destination” = most accurate route planning but requires a specific route and external temperature.
> Using surface street numbers for that is....unhelpful, at best.
Strong disagree here. When doing a trip we always use the nav computer so we always have the real number. (If only so the kids can look at the computer to see the number rather than asking "how much longer" continuously).
OTOH, in town we generally don't, so the top number being accurate for in town driving is very helpful.
