NYC introduced fees for ride sharing with destinations or pickups in Manhattan to reduce congestion. They also added fees to guarantee a reasonable minimum wage for the drivers of the ride sharing app.
The difference in price between a cab and a uber/lyft is relatively small. Convenience seems to be the primary driver for which people use (cabs in Manhattan frequently because of ubiquity, apps in other boroughs).
No, this absolutely checks with my experience - I chase 5 and 10ms improvements all the time because we've measured and know it increases conversion.
But it makes sense, too: if the metric here is average latency, that doesn't mean that some users didn't see a much more dramatic increase. Every tiny bit of frustration removed from the experience adds up.
The latencies stack up for any web page. It's not like "oh this file was served 5ms sooner", it's an accumulation of latencies over all assets, and the interactions, that are required to present an experience to a user.
Also, latency measured at the server is amplified once its received by the browser. And when the user's connection isn't great, all this is worsened. It quickly adds up. In fact, it doesn't "adds" up, it "multiplies" up.
Most pages nowadays are spamming requests to backend services, for business logic or not, that's just reality. Barely any actual real world money making website is going to have < dozen requests per page, and won't need a user initiated page load every few seconds to minutes.
A 10ms average improvement could mean 1 in 1000 customers went from 10s to 1s without any other change to other customers.
This is easily possible if you have a highly distributed customer base, and/or some small segment of your customers don't have good upstream peering with your provider.
>A 10ms average improvement could mean 1 in 1000 customers went from 10s to 1s without any other change to other customers.
Which is why I asked the question in the way that I did. I buy that a slimmed down webpage loading 10 ms faster on average will increase conversions because that makes the site usable for the visitors on bad connections. Moving to a CDN doesn't have that impact. It shaves off 10-100ms across the board.
> Moving to a CDN doesn't have that impact. It shaves off 10-100ms across the board.
I think this is where we disagree. I've seen (firsthand and through analytics) situations where using a CDN can dramatically improve response time in a small subset of customers (while also getting the across the board win for most customers).
I've also seen CDNs (Amazon's in the early days) that were signficantly slower than direct to linode, even with a warm cache. It's a weird world, and packet routing is hard.
What kind of storage do you use when chasing the 5-10ms improvements? I ask as someone working on super fast storage and memory and it's not always easy to find people who are aware of the differences in latency. I imagine you're already on NVMe SSDs, but if not, I'd be curious why (i.e. is software more of the issue than hardware?).
If an airplane crashes because it's hit by a missile, or because there's a bomb aboard, or because a suicidal pilot flies it into a mountain, that is no reason to ground it.
Thus, more evidence is required to ground an airplane (that has, after all, gone through the certification process), than two crashes.
I agree with the assessment that the MAX has a problem; I'm saying "oh, two crashes" by itself is not sufficient evidence for that conclusion.
As someone pointed out in this thread, there were three B767 hull losses between September and November 2001, but two of those were caused by 9/11 terrorists and don't tell you much about the airworthiness of the aircraft.
FWIW, IMHO it's not pedantry, it's precisely the core of the discussion. The 737 MAX was deemed airworthy during certification, and deemed airworthy after Lion Air, and a further crash by itself does not change that. It is only the details and circumstances of the second crash that can provide evidence to challenge the conclusion of airworthiness. Just reflexively saying "crash, ground it all" is mistaken.
The point I'm trying to make all along is that it is precisely "those details and circumstances of the second crash" (that have emerged over the last few days) that make it so worrisome and warrant a grounding. It is not the fact of a second crash by itself.
The fundamental question is this: should the plane have been grounded immediately after the second crash? Many here seem to think that yes, obviously. I think that no, one unexplained crash of a certified plane does not warrant grounding. Once details and circumstances emerge that are indicative of some fundamental design flaw, or multiple unexplained crashes (as with the comet), of course ground it.
Is that a substantive disagreement, or "pedantry"?
When the second crash occurred, that was then TWO unexplained 100% fatal crashes within 5 months, with extremely similar circumstances and evidence pointing to the same root cause. Basically, the first crash with lion air had them on red alert for this model of plane. The second crash was nearly identical and was all they needed to ground them instantly.
On 24 March 2015, Germanwings Flight 9525, using an Airbus A320, flying from Barcelona to Düsseldorf crashed near Digne in the Southern French Alps, killing all 150 on board. They didn't know it was a suicide pilot until after they investigated. Yet, A320s kept flying in the meantime.
On 29 March 2015, Air Canada Flight 624, using an Airbus A320, flying from Toronto to Halifax carrying 138 people crash landed short of the runway. The aircraft was badly damaged and 23 people were injured.
On 14 April 2015, Asiana Airlines Flight 162, an Airbus A320 with 82 people on board, lost height on final approach to Hiroshima Airport in Mihara, Japan, struck an antenna, and skidded onto the runway on its tail, spinning 180 degrees before coming to a stop. Its main landing gear collapsed and the aircraft suffered damage to its left wing and left engine. 27 of the 82 people on board were injured.
Three incidents, on respectable, high-safety airlines within 3 weeks of each other all with the Airbus A320. Yet, those planes weren't grounded, not even for five minutes. Three serious incidents within 3 weeks of each other is "empirical evidence" that the A320 is unsafe.
But we have a crash from Ethiopian Air, on a plane where the first officer had only 200 hours of total flight experience. And another crash months ago on Lion Air (a ridiculously unsafe airline) and that's "empirical evidence" that the airplane is bad?
Lion Air should be grounded. Ethiopia shouldn't be flying planes with student pilots in the right seat. There may be issues with the 8 Max, however, it's a fact that inexperienced pilots, bad maintenance, inferior safety processes and negligence will exacerbate any potential flaws in the aircraft. The fact is that in the case of Lion Air, if it were Southwest Airlines operating those flights in Indonesia, flight 610 never would have crashed. Lion Air had days of reported issues before the crash and they ignored them. Look at Lion Air's incidents and accidents: https://en.wikipedia.org/wiki/Lion_Air#Incidents_and_acciden... It seems that can't even taxi a 737 safety, let alone be trusted to fly one. And right now we're complaining about the FAA? We should be complaining about airlines that are being allowed to operate with this slipshod respect for safety.
The US, Canada and Europe hasn't had any 8 Max incidents -- the incidents happened in organizations that were operating in ways that would never be allowed in US, Canada, or Europe. That fact seems to be getting ignored in favor of complaining about the FAA or Boeing. The US flies a lot more 8 Maxes than Ethiopia or Lion Air, yet not a single crash. Is that just coincidence or does the US/Canada/Europe just do a better job?
Asiana Airlines Flight 162 was an Airbus A320-232, manufactured in 2007
Air Canada Flight 624 was an Airbus A320-211 manufactured in 1991
Germanwings Flight 9525 was an Airbus A320-211 manufactured in 1990.
The Asiana and Air Canada incidents were similar, but the aircraft were of different models, one was 7 year old and the other was 24 year old. The German Wings and the Air Canada aircraft were similar models, but both planes were 23-24 year old, and the incidents were completely different.
Compare that to the 737 MAX 8. Two fatal crashes, during the same stage of flight, with identical models, of similar age. 2 out of 350 planes crashed, where the fleet is barely one year old on average.
I read through the codebase and can't assert that couldn't happen... wasn't sure if the ngx_http_ssl_certificate callback could be executed after a point where any of the client-controlled variables from [1] are defined.
I wish there were some public data on the impact of removing those - app start latency, battery consumption, bandwidth, etc. I bet there are positive impacts for the customers other than just privacy.
Not necessarily.
Fevers can cause seizures and they can have detrimental effects in paediatric patients. For the bulk of the general population, these painkillers/anti inflammatories are simply symptom relief.
For those talking about hyperthermia killing, yes but in almost all those cases the cause is not an infection.
In fact, I would say that paracetamol kills a hell of a lot more patients than it actually ‘saves’.
A study I have taken to heart is that permissive hyperthermia (up to 40 deg c) in ICU patients has a greater survival than those where fever is treated aggressively.
Physiologically, this resonates because high temperatures activate the immune system and raised temperatures are non-optimal for bacterial proliferation; so the immune system is primed by fever; suppressing it can dull immune system response.
In fact, malaria was used as a treatment for syphillis in the early parts of the 20th century because high temperatures kill spirochetes. There are also a decent number of case reports of cancers going into remission following fever.
However in my quick mobile google then I could only see the following study that demonstrated no advantage for either control of permissive hyperthermia group in ICU patients; so perhaps I was relying on a study that has been superceeded.
Fever can indeed kill. Raise the body temp a few degrees, no matter the cause, and things start going wrong very quickly. Google hyperthermia.
In fact, i cannot think of any disease that literally kills. Even with the big stuff like cancer or aids, it is always the symptoms that get you. They damage body systems and the decline of those systems (aka symptoms) eventually causes the cardiac arrest or internal bleeds that shut off nutrients to the brain. Those symptoms are just as lethal no matter thier cause. A massive fever that stops normal body chemistry, whether caused by flu or ebola, will kill you just the same.
Yes but the extra heat in the body causes the same cascade of chemical changes that will eventually kill. Initial treatment is different, to a point. If a fever is too high it will suddenly need the same treatment as hyperthermia: direct cooling.
Not a doctor: Isn’t fever (objectively, it seems) defined as a rise in body temperature? From what I’ve learned about protein denaturalization in high school, that doesn’t sound good if excessive.
