I had to look this up. In California in particular, this is true, which surprised me.
Per a random law firm:
California’s yellow light law permits drivers to enter an intersection during a yellow light. No violation exists unless any part of the car is over the stopping line when the light turns red. However, the law encourages drivers to slow down before reaching the intersection.
Whereas in, for example, Massachusetts, this would be considered running a red light.
If this surprises you, consider the alternative: Driving through a yellow being illegal or unexpected doesn't make sense, given the finite stopping distance of cars, and reaction time of humans. This is because the yellow light is the first explicit indicator you must stop.
If this doesn't make sense still, picture this scenario: You are driving at the speed limit. You are 500ms from crossing the stop line threshold. The light turns yellow.
Your interpretation would make sense only if there were a (paler yellow?) light warning of the yellow light!
Yep, something something Chesterton's fence. If you couldn't drive through a yellow then you effectively have two lights go/stop which makes drivers choose between the safe but illegal thing of running the light or the dangerous but legal thing of slamming on their brakes.
You can't have an instant switch between go/stop which yellow— effectively meaning unsafe stopping distance go, safe stopping distance stop— solves very neatly.
In France, it is illegal to drive through a yellow light unless it would be dangerous to stop, for example if it required you to break hard. That part is up to the officer judgment.
In practice, I have never seen it applied, and it is only a small fine anyways, much less serious than running a red light. I guess it can be used as an excuse if the police really wants to pull you over.
So a pre-yellow light? This seems like you just changed what we call a yellow light to the 4 flashing greens, and made the yellow the new red. If yellow means “don’t enter the intersection”, how is that different than a red?
The US sometimes has something similar: visibile countdown timers for the pedestrian crossing turning form stop to go, which coincides with the car light turning from red/stop to green/go.
This encourages people to run the light by trying to turn exactly as the countdown timer hits 0, trying to race against pedestrians trying to cross crossing pedestrians.
You could always do that before in most instances just by watching for the yellow on the cross street too. Though I think the green is often slightly delayed relative to the pedestrian light, precisely to ensure the car cannot win that race legally
On the contrary, I think it could make perfect sense: What's written in the law is one thing, and how it's enforced is another; I would argue that the former should be done with consideration to the latter.
Based on my own experience, I'd estimate that well over 99% of traffic infractions go entirely ignored by the law: minor speed violations, unsignaled lane changes, rolling stops at stop signs, expired tags, cell phone usage, and yes, running red lights.
When the letter of the law is broader in scope or errs on the side of caution, that enables the police to exercise their judgment in enforcing it (with the obvious caveat that some police will abuse any power you give them). You could imagine a scenario where someone technically runs a red light but it's totally justifiable and safe (heavy load + moderate speed + short yellow + no other traffic) and another where someone technically makes it into the intersection on a yellow light but senselessly severely endangers public safety (busy intersection + rapid acceleration + traffic backed up on the other side of the light).
I would be okay with someone evading a ticket in the first case and getting one in the latter.
In the UK, if you're at all over the line when the lights change, you're considered "in the junction" and are expected to leave the junction -- the next phase should give you priority to do so. The only way to run a red light is to start crossing the line while the light is red -- although plenty of drivers will start to inch across while they're supposed to be waiting :P.
The most annoying scenario is where a driver has either stopped or inched forwards far enough that they can't actually see the lights any more and don't know when they've changed.
This is the law in Illinois, too. It, in combination with the way the lights are timed, makes some intersections particularly challenging for pedestrians.
The basic way the timing goes is: traffic light changes, pedestrian crossing signal illuminates, traffic going straight that squeezed in on the yellow finally clears the intersection, cars turning left finally get a chance to go, pedestrian can finally safely enter the intersection with approximately 10 seconds left to cross a four lane street, lights change again, cars start honking at the older person with mobility issues who could only get halfway across the intersection in the time they had available to safely do it, impatient driver from further back in the line who doesn't care to figure out why the person in front hasn't started moving even though the light has been green for five full seconds swerves into the right turn lane and guns it, narrowly missing the aforementioned older person with mobility issues as they blast through the intersection.
I'm wondering how you would know when the yellow light was going to come on.
Do you have some sort of countdown, or innate knowledge?
Because, otherwise do you just randomly stop at green lights guessing that a yellow light might come on? Or do you drive so slowly that you can stop in the width of the white line before a pedestrian crossing? Really, I'm trying to figure out how you don't ever enter just as a light turns yellow. Once you do, do you stop in the intersection or try to clear it before it turns red? I hope the latter.
For me yellow lights are a warning that a red light is coming. It should be long enough for cars to clear the intersection (in many states without gridlock rules even this is not the case for left hand turns).
My experience in Boston is that drivers try to beat the green light change and accelerate while it's still red.
> For me yellow lights are a warning that a red light is coming
Correct, yellow means "start slowing and stop before the intersection if you can do so safely, otherwise proceed". Red means "do not proceed if you aren't already in the intersection".
This is why the opposing traffic signal and walk signal will wait for a second or two after red: to allow people who entered on yellow to finish their transit across the intersection.
Indeed, in the author's own video where they incorrectly claim someone ran a red light, the author had no legal right to cross anyways, so there was no chance of the author getting injured unless they ran a red light at the crosswalk.
In short, the author seems most frustrated that the situation changed from everyone waiting on him, to him waiting a few seconds for others.
I assume you mean the perpendicular direction crossing lights?
Most pedestrian lights have delays longer than that in California for safety so you might stop at a green light.
Also, if nobody presses a button ped-lights don't even turn on, just like left turn signals don't turn on without a vehicle triggering it.
> Also, if nobody presses a button ped-lights don't even turn on
This is no longer true in many cities. Most SF crosswalks don’t require a press anymore, many don’t in LA, and all don’t in New York. AFAIK it was a Covid thing, back when people thought surfaces spread the virus, but it stuck.
It’s unlikely they are able to gouge McDonald’s and the large companies. Even the article mentioned that the big competition is for those massive contracts. It’s much more likely that McDonald’s and co can shop around, negotiate a great price, and maintain margins.
It’s the mom and pops, the regional suppliers that can’t do anything, and likely pay much higher prices than the megacorps.
...despite being started at the same time in 15 different locations by humans, in the most windy and hot night of the year, the wildfire got surprisingly big
As you consider the decision to let them go, consider the impact on other people in your work place too. Other developers notice someone like that underperforming and getting overpaid, and it can hurt the good climate you aim for. It may feel bad to fire them, but it may be worse to keep them.
You absolutely need to consider how the rest of the team will interpret you actions. Most probably know this individual is underperforming already. Teammates usually have a sense before the manager does.
But the way you handle it will also be observed. Others need to know you won’t just fire someone on the spot if performance doesn’t match expectations. Life happens, people have personal life issues that may occasionally require attention, and you don’t want everyone else assuming they are one bad sprint away from unemployment.
In most of the US at least, start with a performance improvement plan. Ideally one crafted with HR’s help. It should lay out specific expectations for a Senior Engineer, and call for immediate and sustained improvement. Use that to initiate a conversation and clearly lay out that they are not meeting expectations for the level at which they were hired. Present meeting those goals as a requirement for continued employment. You may also present a severance option here if you want them to have an out.
This makes it clear to everyone that you treat all employees with respect, offer a clear warning, and then so long as you expectations in the PIP are clear, it’s a lot cleaner if you end up needing to part ways a week or four weeks later.
As a counter point to this, I’ve witnessed the practice of designing these processes to terminate someone creates a ton of burnout for the manager. Expect to spend at least 20% of your time managing the PIP. Documentation, one on ones, babying tickets being assigned, etc.
On top of that I’ve never seen anyone get through a PIP successfully (n=6).
Furthermore, it can also create a culture of just dismissing any candidates who don’t come from the right pedigree because the risk of a mishire is way too high.
Its worth asking if it’s worth your time to do this for the sake of following process. Larger companies that need to standardize management for hundreds of managers have a greater need than startups do.
> I’ve never seen anyone get through a PIP successfully (n=6).
I don’t think they’re meant to succeed. If you want to keep an employee, you give them feedback and coach them. A PIP is used to document evidence that an employee is incompetent and is incapable of improving despite your best efforts, as a precursor to firing them.
It’s definitely not zero effort. But every time I’ve seen it draw out the timeline, it’s been because the expectations weren’t clear at hire time, nor in the PIP. You’re looking for a specific number of “significant PRs/week”, or story points, or something with subjectivity where you the manager still get to decide.
It’s a warning shot, and sometimes it only lasts a week.
BigCo HR can definitely make it take months, but the practice in general doesn’t have to be that way. We did this at startup size as small as 20 people, and I never regretted it.
Do you have thoughts on offering PIP vs demotion and salary adjustment?
If the individual doesn't have the skills (or seemingly temperament) for a senior role, I'm not sure PIP is setting them up for success.
Versus demotion with possibility of promotion allows them to perform at their level of competence, leaving open the possibility for them to chose (or not) to aim for promotion.
I haven’t personally seen that work in practice. I’ve been in situations where we considered it, but people usually have picked a role with salary considerations in mind, and often can’t really walk that back, and if they do it’s demotivating. And they may be able to perform elsewhere in a way that meets the need at your senior comp level.
I think it’s easiest to cut the relationship and let them start clean somewhere else.
This 100%. Whether I would fire someone depends on why I thought they were underperforming. You should consider the first month on the job a work sample and if someone isn’t measuring up within the first week tell them what your expectations are and that they’re not being met. If they don’t remediate within a month let them go.
I kept someone for too long once and people on an adjacent team didn’t want to get stuck with that person during a re-org and were vocal about it. And that person bungled some important projects that set us back years on our ambitions.
Fire them and use this as an opportunity to address your interview process. The longer you keep them the worse it will be.
How do you reconcile this advice with the fact that in both places I've worked at, the expectation was that just the onboarding takes significant amount of time (up to ~0.5yr)? Even if I personally didn't need that much time, it was nice to know that it's available -it's not easy to get into the grind and catch up to others that already have all the institutional knowledge when starting a new job.
You shouldn’t need to be onboarded to basic development skills when you come in as a senior developer as the OP is saying here. Institutional knowledge specific to each company takes time to learn, but many dev skills should be directly transferable.
Be sure that half a year is thought through. I’ve seen places that pay too little attention to new employees in the first few months and then surprise them with complaints about performance. Those first few months should be drilling specific institutional knowledge (through bugs, presenting on parts of the stack, talking to a variety of people, shadowing, studying certain tech/tools most people don’t know.)
You should be looking for the employee to be making good general efforts, just without taking everything into account yet. And the employee should see evidence that you are serious about helping them master specifics of their new environment.
From this post there's not really any indication they're a "bad employee" though. They aren't up to the needs of the team that hired them, but that was on that team to figure out before they hired them.
It's probably fine to fire them, but also everyone involved should be open about the fact that the hiring and so also the firing is the direct result of their own failure. The morale implications of this could be lot more nuanced than you're implying.
If I were working on that team, I wouldn’t like the company firing the senior dev. If they keep them, and train them, that would reinforce the idea that the company is taking care of its employees. That’s nice.
I hate this rat race of being a performant engineer. The moment your performance declines, your managers will let you know and you’re under the radar now.
> If they keep them, and train them, that would reinforce the idea that the company is taking care of its employees. That’s nice
I don't think so. A company is under no obligation to handhold their employees. It should be a 2 way street, beneficial in both directions. If it's not beneficial for one party, they should be able to be honest about it and move in another direction. We're all adults here.
You are right in general, but in this case it is clear that the issue is not "guy has a tough break, going through a divorce, underperforming for a little while - let's wait it out". It's "we thought we got Messi and we ended up with Henry Bemis - which cannot be salvaged. Imagine the other top players on the team seeing Henry Bemis getting paid Messi-money.
This article is just an exercise in back of the envelope math. There’s no coherent point to it. “If we just assume that Pablo Alto was as desirable as Manhattan and could justify building the same infrastructure…”
Great summary. Always hard to tell whether proposed experiments are doable within a reasonable timeframe. Sounds like this one is not, but perhaps by defining a new category it may lead to one that is.
Go to the "Getting Started" page and you will have instruction there on how to set up the Figma library. It's pretty simple. Let me know if that worked for you.
This looks pretty compelling! The best of controlling your own data and a global, network-effects-compatible content approach. Seems like a win for users if it gets adopted.
I'm not sure if it's concentrated in blood or not but as I understand it it's mostly protein bound which means it probably moves into and out of tissues easily but not out the kidneys or into stool.
Someone else's already replied with the Australian fire fighter study which shows that blood levels fall with donation. If movement isn't free into and out of the "compartments" then it's possible end organ damage continues unabated even as blood levels dilute.
But it seems likely that donating does reduce concentrations in organs as the PFAS move back into blood.
Fascinating. This material alone would be revolutionary if legitimate, although I’m sure there’d be further improvements.
Question for any experts - what’s the relative difficulty of keeping something under sustained high pressure in a piece of hardware vs keeping it very cold?
Our ultra cold usages work decently well. Would it be any easier to keep a hardware component under pressure like what this new material requires?
Keeping things cold is effectively an energy vs space tradeoff.
For example, if you wanted to keep an underground superconducting wire cold, then you would send coolant pipes along it, and wrap it in an insulator. You need to put energy into chilling the coolant, inversely proportional to the thickness of the insulation.
However, typically for most things humans want to do, the cooling cost works out higher than the energy lost to resistance in a non-superconductor, so, apart from a few specialist use cases (MRI machines, particle accelerators), superconductors have seen no use.
The second to holy grail is temperature above liquid nitrogen at around normal pressure with cheap and easy to obtain coolant. It will enable big things like really high voltage intercontinental transmission lines etc.
We've already got there with YbCO. But just barely, which means useful amounts of current and magnetic field density bring it out of the critical region.
Is there some napkin math available on the net for a transmission line with a nitrogen-cooled high-temp superconductor (Tc > 90K) and thick thermal insulation? I mean for the energy required per km too keep it cooled below Tc.
Big high voltage transmission lines lose ~200 watts per meter in resistive losses when under full load.
The electrical energy to keep something 1 meter long at liquid nitrogen temperatures is also ~200 watts, assuming 8 inches of insulation.
The resistive losses go down with the square of the power transmitted - so they fall to zero rapidly when not under full load. Cooling losses stay approximately constant.
Therefore, I suspect a liquid nitrogen cooled superconducting cable wouldn't work out financially.
This is my own ignorance, but what determines the power carrying capacity of a wire besides melting from the resistance? Could you transmit a lot more wattage through that same line when superconducting?
The energy loss (in the sense of where the heat ends up) is at the refrigeration plant. The cable itself extracts heat from the environment. So SC cables make sense for underground cables, where heat buildup is a problem.
Per a random law firm: California’s yellow light law permits drivers to enter an intersection during a yellow light. No violation exists unless any part of the car is over the stopping line when the light turns red. However, the law encourages drivers to slow down before reaching the intersection.
Whereas in, for example, Massachusetts, this would be considered running a red light.
https://www.wccbc.com/red-and-yellow-light-accidents/#:~:tex....
reply