I'd like to provide an personal take on this - this will be a bad deal until Uber can be a super app doing 100x more things than it's doing today. So maybe never.
Why? Because exactly in China there was consolidation of food delivery apps into like 3 or 4 of them from 100+. But all of them are still burning lots of money from investors, except one - Meituan-Dianpin which built on this super app (first think of it as uber+yelp+groupon+tripadvisor+more) handling all kinds of life needs from food to labor service. Yet this super app barely started making profits recently after grabbing so much aspects of daily life, while its food delivery unit is still burning money, though contributing to the profiting units like ads.
The American equivalent would be Google, as for myself and many others the Google maps reviews have long replaced yelp/groupon/tripadvisor, as it perhaps surprisingly tends to have the most accurate information about local businesses. But Google does not get into delivery because in America, the math doesn't work.
When I was in HK, I used foodpanda, it was super cheap, I think $10HK. Deliveryman shows up on a scooter, and has at least one other persons food with him. The density supports it. American cities are built somewhat differently where most middle-class people don't live very close to most of the restaurants they wish to order food from.
I think it has much more to do with the terrible math involved with doing delivery in such spread out regions. There being a super-app that does everything does not matter.
Also in America food takes much longer to be ready than in Asia. I don't know why but it does add up.
The title misses the other, and more impactful/unprecedented change of giving everyone "exceeds" rating and skipping perf review for them (except for those high performers who will get separate review).
That does far more than $1000 to relieve everyone and make it fair for those who have to take care of others in the family during such time.
Yeah, this change is easily worth something on the order of $20k this year and then $10k per year for the three following years for every engineer who wouldn't have otherwise achieved this rating. It's huge.
Interestingly, I assume they're doing this because it means they can skip the performance cycle entirely. That's a good way to save process. It's gonna harm some people who were on track to do better than exceeds, though. (On average it should help people.)
> It's gonna harm some people who were on track to do better than exceeds, though.
There's no news about this yet, but I've heard that people that are on track to do better than exceeds will likely still have a review so that they don't miss out.
Minor correction, China REPORTS making rather good progress.
While I am neither a tinfoil hat wearer nor have anything against China, the fact that they were silencing a Wuhan doctor who was trying to alert people and authorities to the virus all the way back in December makes me skeptical of their reported status.
China isn't an information blackhole nor do they have perfect censorship. If the outbreak were still accelerating in an exponential fashion it'd be impossible to hide it from the world. They have successfully hit the inflection point on their logistics curve of viral spread (for now, anyway).
And China is reportedly sending equipment to Italy - not something they would do if the infections were still growing. Their actions seem to align with their message.
Genuine question, do you happen to have a source? Not that I don't believe you, I am just curious about the details, such as how many units, for how much (is it free?), etc.
> Wang assured Di Maio that Italian orders for the ventilators would be made a priority by Chinese firms after similar requests were made by other European countries.
Thanks for sharing, I really appreciate it.
Please correct me if I misunderstood, but the wording makes it sound like they were ordering them from China, i.e., buying, not receiving it as a donation. Which is exactly what I was afraid of. Money is money, and I can totally see some governments, especially the PRC one, prioritizing that over the actual help for their own citizens. Especially in the scenario where they know they are screwed no matter what, so might as well make some quick bucks.
Yeah, it's a story of failed censorship and the truth coming out anyway, not successful censorship hiding the scale of the problem. That's impossible so long as China is on the Internet.
They cannot stop all of it, so things will definitely leak out. However, they can stop a giant majority of it, so while we still get the leaks, the magnitude of the prevented leaks is probably still significant enough that the censorship can be considered as "working".
Do not underestimate the Great Firewall. But anyways, from a manufacturing and distribution of basic essential health supplies, and more, China is probably best positioned to help save the world.
The Wuhan authorities tried to silence her. There's no real reason to disbelieve their numbers now, beyond the problems of getting good numbers in a large country.
Read the WHO/China report. Or look at South Korea.
Many people here are conflating "stop" with "make it non-exponential". "Stop" just ain't going to happen, and hasn't happened anywhere (Vietnam reports they have, but come on). China reports bending the curve away from exponential. That is the goal. It is a very important one.
It's not just no longer exponential. The number of active cases is China is in decline. How long they can keep that going is debatable. But as of right now they are effectively stopping it.
That is not stopping. Extrapolating your trend to infinity suggests stopping is plausible. Latest data I see is 134 new cases per day in China. Let's take those numbers at face value and believe them (no sarcasm intended). However, these are just known cases. China is not testing everyone. As managing the spread becomes more successful it becomes cost and manpower prohibitive to test enough people to catch them before they spread significantly.
China has moved the R0 below 1 through some combination of behavior changes, else the number of active cases wouldn't be going down.
If they can keep it below 1, it will die out. It's possible there are many more unknown cases, but if unknown cases are increasing, confirmed cases should be too.
What I would encourage people to consider is whether this virus is at the level that we would all be happy to allow military enforced city-wide quarantines or travel restrictions...
I understand that this is NOT the flu, but it seems like the best data we have puts the most pessimistic CFR at about 0.6% if you look at the South Korean data (who have done, by far, the best job testing en masse).
I agree that slowing the spread of the virus to help our health care workers avoid being inundated with admissions to the ICU is worth while, but I'm extremely skeptical of embracing what China has done.
> but it seems like the best data we have puts the most pessimistic CFR at about 0.6% if you look at the South Korean data
Check your stats. 0.77% of South Koreans who tested positive for the virus have died already, and more of them will die in the future. The CFR is likely to be above 1%.
Right now the best information I've heard from several epidemiologists who study pandemics is that we are still seeing the results of severity bias, in that people with absolutely no symptoms will be far less likely to get tested. Also, South Korea is not seeing anything close to an exponential increase in cases.
When those same experts see numbers coming out of Italy, they believe very strongly that the number of infected is far higher than being reported because testing is not as widespread.
~0.5% is a very reasonable estimate of the initial CFR when an epidemic starts. It's bad, around 5x as much as the flu, but probably not worth a national China-style quarantine.
But it has a key context: that's the CFR for when all patients are treated. In most of China, there were never enough infections to trigger a healthcare system collapse: that's why in all but one province, the CFR was indeed ~0.4%.
In one province, Hubei, the medical system did collapse.
Its CFR jumped to 3-4%. That is worth going to extreme measures to prevent. In the USA, that would amount to ~10M dead.
That is not a seasonal flu.
I think a superior approach for most of the world is one along the lines of Singapore, Taiwan, and South Korea: extensive testing and contact tracing and limited but real quarantine measures. But, yes, everything should be on the table to prevent the nightmare healthcare system collapse scenario.
CFR of 0.6% is a very optimistic figure. It is a naive calculation based on current # deaths / # infections.
1) But # infections were growing exponentially, we need to use numbers from the same cohort, which implies much lower # infections & higher CFR. Naive CFR will go up once infections grow more slowly. (It already is higher than 0.6%).
2) South Korean confirmed cases are much younger than their median age, mainly between 20-29 years old (perhaps because of where superspreading events happen—that church). This age group has a much lower fatality rate from Covid-19.
3) # hospital beds per capita in South Korea is second highest among OECD countries (1st is Japan) and ~4 times that of the US. They already have patients waiting for beds. Most countries will do much worse if they reach the same # infections per capita.
We can embrace what China has done, or what S. Korea has done, or what Singapore has done, or what Italy has finally started doing. Any of those would have a dramatically positive effect. Right now, we (U.S.) haven't done any of that. That is the problem.
China is now seeing a wave of secondary infections as they relax travel restrictions and as people enter the country from elsewhere. They have not stopped the spread, they simply slowed it down.
That's the whole point. If you slow it down you can avoid the rapid spike in severe cases that overwhelm the system. It's like traffic -- once you surpass a certain volume, the system locks up and throughput drops.
I think the point was to do a controlled restart while making sure the rate of secondary infections remains low enough that the impact of the coronavirus remains less than the flu.
How do we know that the numbers are trustworthy? They weren't for the first two months and China has clamped down on any information leakage. Meanwhile as of last week their factories were still at around 50% capacity or less (check pollution maps online) and now I'm hearing rumors that they're forcing uighurs to work the factories that migrants are refusing to go back to.
I'm not convinced yet. This would not be the first time a communist authoritarian regime lied about a bad situation - far from it, it's standard if you look at history in the Soviet Union for example. Authoritarian regimes cannot work without being respected and/or feared by the people, so they are incentivesed to lie to save face and simultaneously punish anyone who questions the lies.
Especially considering that the longer this drags on, the more likely nations and corporations alike are to pull production from China permanently.
I’ve also read this book a couple times but sadly as a engineer in large and mature company I hardly find any opportunities to actually apply or strengthen the learning, then as time goes my learning just decays.
How are you tackling the issue? It’s not like that one can find distributed system projects easily.
This book is on my reading list, but I haven’t gotten to it yet. However, I would like to share my experience, since I was in a similar position as you. At my previous company, I was a project engineer (similar to a TPM) and acts as the technical sidekick to the Program Manager. I provide engineering assessment and recommendation to the PM, and oversaw interdisciplinary engineering teams to ensure they are on track for the project. Most of the time the work was not intellectually stimulating, and there was not much room for innovation since the contract, scope, and waterfall schedule pretty much dictated the entire project flow.
What I did, was to seek projects outside of work. In my case, I self-studied python, algorithms, and data structures. I started to implement my own projects, inspired by 12 in 12 challenge by Mackenzie Child. Somehow, I was able to get an interview with Google through the Google Foobar challenge, and I am now working at Google as a SWE.
Like what I shared, I haven't read the book yet, but if I was to apply learnings in distributed systems, I would build one using virtual machines or single thread processes with message passing via TCP, UDP, or HTTP, each representing a system for a small scale prototype. After I am comfortable with the concepts and implementation, I’ll probably scale up using one of the cloud service providers (AWS, GCloud, Azure, Digital Ocean, etc).
At my previous position I was growing project management skills but felt my engineering skills was stagnating. I decided learning Python on the side. Based on HN recommendation, I started reading the CLRS algorithms book and implemented all data structures and algorithm with Python up until red-black tree. After that, due to time constraints I only implemented select algorithms from the graphs section.
After I had a better foundation on data structures and algorithms, I used leetcode and hacker rank as a test bed for my implementations. It is great, since the submission tests cover correctness, edge cases, memory and time constraints.
During this journey, I somehow triggered Google Foobar challenge with my Python and algorithm searches. This was the turning point—-I had tried to apply to Google before, but even with referrals it went nowhere due to my background not matching what G was looking for. Foobar, if you didn’t know, is a timed online programming challenge similar to leetcode. It consists of 5 levels, and Google will give you a phone interview if you completed level 3. This allowed me to bypass the HR filter!
So I started prepping for the interview in anticipation. I did the usual study routine—-cracking the coding interview and lots of leetcode. I wrote a lot of code on paper, completely filling up 2 notebooks. My goal was to be able to pick a random leetcode problem, handwrite the solution in 20 minutes, and have it pass all submission tests. Eventually I was able to do that for most problems at easy/medium difficulty.
The rest is just going through the interview process. Phone, then on-site, the nervously await the result. And you know the rest.
The whole journey took about 1.5 years, starting from learning python and reading CLRS. I know there’s a lot of dislike for the whiteboard algorithm interview, but in my case it was a way to pivot to SWE despite not having the background and experience, and I am grateful for it.
You are a true inspiration for every developer. In the online programming forums such as HN and Reddit you see lots of entitled people who keep saying how poor the whiteboard interview process is. These people are mostly bitter they cannot compete with self-motivated, hardworking and talented people like you. You are the exact kind of developer Google and other big tech companies are looking for. I have no doubt you'll have a great career as a SWE.
I have switched reasonably frequently in the last 5 years (approx 1.5 years at each job). Jobs are never as good as promised and sometimes turn put badly enough that you want to switch again soon. And its always a hassle switching. It does seem to be the most reliable way of getting a pay increase.
This is indeed very interesting to see and made me want to share my stories. I've also scanned over thousands of books (mostly Japanese manga, in a way to digitalize them and make them persist) myself and ever wondered where tech could bring this digitalization to. I've spent more than 10,000 hours on this for more than a decade and will probably do it until I couldn't.
Due to the requirement for maximum restoration of manga contents, especially for those cross-spine scenes, the high speed camera-based solution like above or other places which focus on formal prints won't work quite well and that leads to two major ways in the trade for decades since some Japanese started this self-scanning thing around 80s:
1) Destructively break the book into unglued pages (which also invovle cutting the glue with heavy-duty paper cutter) and send them to scanner with auto feeder, (then try to glue it back if keeping the book is vital, at best effort).
2) Manually press each page on flatbed scanner (thus A3 scanner works better as it allows vertically scanning spine area usually with the greatest shadow) and scan each page using predefined rough area which would therefore include non-content area as shown in : https://imgur.com/a/2ITBlJg (left/right edge and spine).
Solution 1) works great in terms of efficiency but could be a pain for those who love to keep book still in the collection (though malformed anyway, and I myself might consider this when I get old and couldn't bear the effort). 2) would cost a lot more time (on average, #1 takes less than 30 mins per book and #2 takes at least an hour) especially for manually removing noises as shown above. I've tried some basic CV tools/scripts to auto cut at least the edges (middle spine could be troublesome and I'm fine with manually working on that), which work really poorly since content itself could also be largely variant if we try to determine "edge" by checking pixel distribution.
Hopefully before 2nd decade finishes, I either give in to destruction or find a perfect way to automate the noise-clearing process.
22% (6% answered not sure). But to be clear, 22% SAY they know what it is. I guarantee that if asked to explain what it actually is, some chunk of that 22% would get it wrong. Anecdotally, most people I talk to IRL believe it will either lower their cable bill or remove their bandwidth cap.
I think the counter one liner to that is something like: Net Neutrality keeps big companies from controlling what you can access on the internet. But not totally comfortable with that because "access" is overstating.
Good point, I think this is important to highlight, and might appeal to people reluctant to support net neutrality.
If one supports competition and free markets, recognize that the ISP business is not free and open to competition, and used regulation to gain monopoly or duopoly status. We need some regulation to either try and create competition, or at least prevent abuse of their current monopoly/duopoly status.
I have to say it's a typical pattern for rich and connected kids in China to gain success. Top local college => top us college => join GS or whatever similar => trendy founder/CXO/etc. => Now it's their turn to have someone write articles to let ordinary people know they are not just rich, pretty/handsome, well-connected, but also they are super hard working, smart/intelligent, wise to earn their crowns all by themselves. We all know for well-connected in China, there are way more gates open to them. And the stake-holders in the western world can help them open other gates. As I said there is a long history of them doing all this kind of thing. Most people in China with enough experience in that society know that.
It's just a capital-intensive pretend play. They are the players and you have to watch them show off how good they played through mass media and your news feed. If there was an earthquake in that circle, new batch of players would show up.
I forgot to mention the difference leaves different room for how much impact other factors can contribute to the final outcome, which, in my opinion, is more important.
I've lived in different cities in China and US. 4-6 RMB to 1 USD in major cities like Beijing/Shanghai, and 1-3 RMB to 1 USD in less developed cities or sururban area. Less developed a city is, less RMB to USD in purchasing power for most domestic goods.
I did ask my current manager about this decision and he did mention that he wants to find a engineering project manager (we have PMs) who could technically manage projects with him so he's not overloaded. This person might become my manager in the future but not immediately (maybe after 1 or 2 years when he's going to assemble his own team).
Exactly! I've read many HN stories about bad interview experience and I really wish to make it an enjoyable one for both me and candidate by staying away from those new-grad leetcode algorithm questions. Thanks for the kind advice!
Why? Because exactly in China there was consolidation of food delivery apps into like 3 or 4 of them from 100+. But all of them are still burning lots of money from investors, except one - Meituan-Dianpin which built on this super app (first think of it as uber+yelp+groupon+tripadvisor+more) handling all kinds of life needs from food to labor service. Yet this super app barely started making profits recently after grabbing so much aspects of daily life, while its food delivery unit is still burning money, though contributing to the profiting units like ads.