I think I got the magic part. You can store all build system binaries in the VFS itself. When any binary gets executed, VFS can return a small sham binary instead that just checks command line arguments, if they match, checks the inputs, and if they match, applies the previous output. If there is any mismatch, it can execute the original binary as usual and make the new output. Easy and no process hacking necessary.
What nobody seems to talk about is that their resulting models are basically garbage. If you look at the last provided confusion matrix, their model is right in about 2/3 of cases when it makes a positive prediction. The actual positives are about 60%. So, any improvement is marginal at best and a far cry from ~90% accuracy you would expect from a model in such a high-stakes scenario. They could have thrown a half of cases out at random and had about the same reduction in case load without introducing any bias into the process.
> What nobody seems to talk about is that their resulting models are basically garbage.
The post does talk about it when it briefly mentions that the goal of building the model (to decrease the number of cases investigated while increasing the rate of finding fraud) wasn't achieved. They don't say any more than that because that's not the point they are making.
Anyway, the project was shelved after a pilot. So your point is entirely false.
I second that. In addition, most of those companies use a portion of their revenue to buy back their shares, pushing their price up. So, the value created worldwide ends up growing the US stock market.
Ths stock market is not considered in GDP. So by that measure it does not directly impact "economic growth." I can see an indirect relation that says that US investors (primarily through pensions and 401Ks) are the primary benefactors of a growing US stock market which then translates to more investment and opportunity of US citizens, but that is a pretty long trail to account for our economic path right now.
And then even more specifically - at least in terms of local problems - that pushes up the value of compensation for a lot of people in the SF Bay Area which inflates the cost of land there.
Staying out of whether or not the concentration is a problem at the national/international level, is there any realistic alternative short of massive protectionism a la China to force home-grown tech companies in other parts of the world?
The US has a massive advantage of being the largest economy, having a vast single market, issuing the world's reserve currency, and having unique hubs like the Bay Area attracting the best and brightest. It would be hard to replicate its success elsewhere without having some of the above prerequisites.
>being the largest economy, having a vast single market
America is NOT the largest market the EU is MUCH BIGGER. And it is not "America" that commercialises technology, but a small portion of California called The Valley.
> People and processes you’ll have to manage to achieve SLAs like Amazon’s?
In reality, you can have almost any people and processes. The trick is to put your servers and data in more than one place. If you have uptime of just 99% for a server (~3 days off in a year) and have them in 2 unrelated places, you will get 99.99% uptime. 3 places will give you 6 9's. The only thing that has to be ensured by people and processes is graceful fallback.
Notice how I say uptime and not SLA. SLA just means that you will get a little bit of money back if uptime dips below the SLA level. Oh, and for EC2 it is just 99.95%. So, if you really care about your users, you will engineer your systems to stay up rather than hoping that a third-party provider's SLA will save you.
That assumes the only causes of failure are environmental. I’ve definitely seen plenty of hardware failures but software failures are common, too, and keeping things synchronized is going to require more than “any people and processes” - that’s how you learn your backup has never been tested and database replication stopped working 3 days before the failure.
Not forgetting operational failures due to human mistakes when doing delicate stuff on complex environments, and setting up on-prem infra to work like a hyperscaler does... well, it's not easy.
Well, dedicated servers (for which you can have private cage, or VISA compliance, or ...) are a markup of say 30% over base cost, which is still 1/5th the cost of AWS. And even Hetzner will just deliver Kubernetes clusters these days.
These avoid all of the costs you were talking about.
> In reality, you can have almost any people and processes.
You've never tried it, huh?
The reality is you will need some very specific processes.
You'll want a test environment, so you can make sure that proposed router reconfiguration actually does what it's supposed to do, and a process that says to use the test environment, and a process for keeping it in a consistent enough state that the tests are representative.
You'll want a process to make sure every production change can be reversed, and that an undo procedure has been figured out and tested before deployment. When that's impossible, you'll need careful review.
You'll want a process to make sure configuration changes are made in all three production data centres, avoiding the risk of a distracted employee leaving a change part-way rolled out.
But you can't roll out to all three sites at the same time, what if the change has some typo that breaks it? So you'll want a gradual process.
You'll want to monitor the load on the three systems, to make sure if one goes down that the other two have enough capacity to take over the workload. You'll have to keep monitoring this, to keep ahead of user growth.
Did I mention the user growth? Oh yeah we're expecting a surge in demand just before christmas. The extra servers we got last christmas have absorbed our user growth, so we'll need more. Of course it'll take time to get them racked and set up, and there will be a lead time on getting them delivered, and of course a back-and-forth sales process. So of course we'll have to kick off the server ordering process in August.
Of course, there's a chance of a partial failover. What if the web servers are still working in all data centres, but the SQL server in data centre B has failed, while the replicas in A and C are fine? If there's a software hiccup you'll need to figure out who to call - yet another process...
That is fantastic news if true. AWS and two other major cloud providers have done everything in their power to make it painful for businesses to switch off of them. Case in point: egress data fees are something like 80x compared to what the cloud provider actually pays. You still have to pay them in full unless you decide to leave AWS completely.
Try using Cloudflare R2 with zero egress and you'll rapidly discover there are lots of 'gotchas'...
- Want to use a cloudflare managed domain - pay for egress if you have enterprise account
- Want to use their 'dev' domain - rate limited
It's still a good service - but zero egress comes with conditions. The only exception I've found is Cloudflare pages which seems genuinely zero egress (as long as you don't proxy it through a managed domain).
Those fees are egregious, but they don't make it any harder to switch off AWS. The cost of the monumental amount of engineering time such a task takes is a couple orders of magnitude greater than the cost of the data egress.
It's almost to the point where you wouldn't even bother to factor into a cost calculation.
That doesn't really make sense. AWS is very profitable [1]. That means that whatever the amount of money they put into engineering, their customers pay, and then some.
As for the customers, very few of them operate at a scale that requires the monumental amount of engineering required to replicate the whole of AWS with its many dozens of services.
Setting up a database, backup processes, object storage, archiving, and especially compliance is a huge upfront cost for a small company.
It is true that you pay for the engineering done by AWS, but you also need to take into account the network effect and compatibility with third-parties.
With this being said, AWS is still very expensive. But AWS being profitable doesn't imply it is profitable for you to run your own infrastructure. The same way it is not profitable for you to run your own shipping company, or your own internet backbone provider.
Indeed, that is decided on a case-by-case basis. So if a company determines that moving on-premise is cheaper than AWS, perhaps that is actually true for them.
That entirely depends on how you build your stack.
I always strongly push for setting up the whole infrastructure with a future migration in mind, and without tying the company closely to specific cloud services.
You lose out on some of the better cloud features, but it's worth it in cash alone: being able to say "we have built our stack to be platform agnostic, we can switch within a month" is very useful in discussions with sales...
AWS claiming that on-prem solutions are competition to UK regulators is like Apple insisting the App Store isn't a monopoly to EU regulators, Google asserting that Chrome isn't dominating the web to the DoJ, or Microsoft denying its omnipresence over the workplace. It's laughable. Seriously, did anyone here actually bother to read beyond the headline?
Those are indeed similar situations, but your examples except the first one do not come from the article. So it is not clear why do you assume that everyone did not read beyond the headline.
If you dig into the article, it's evident that AWS is just trying to appease the UK's Competition and Markets Authority. To claim there's no competition would be a monumental blunder, like a government declaring there's no need for a military during a war. They don't really believe that on-prem is serious competition, but they pretend to in order to avoid litigation.
still not legitimate, because then any laying of on union members/leaders could be legitimized that way
in any country which properly enforces labor protection law laying of any union leaders without the agreement of the union is extremely hard and requires missteps of the members like e.g. stealing. Or really unusual situations like you lay of half of the members and over half of the members have young children (or e.g. are disabled, project leaders etc.) but non of the union members have any of that. The likelihood of which is basically 0 in practice.
If that went to an arbitration, the HR would have some difficulty unless they could show a paper trail (eg emails in which people expressed concern about the cultural fit of these individuals) ahead of time. They could of course have paved the way by preparing such a trail ahead of time…
But that would lead the way to the question of why this was a redundancy rather than dismissal for cause.
Im not any sort of lawyer let alone an employment lawyer, but I’m sure there are some employment lawyers getting in touch with these folks now to test their interest in pushing a case.
if these employees were focused on the organization efforts it seems reasonable that it would be pretty easy to show them as subpar performers against the corporate-oriented performance expectations, compared to others who were focused only on their work tasks. Stacked Ranking is reprehensible but not illegal.
Bad culture fit is hard to argue when those 8 people were literally elected by their peers to represent them. One might say they're as perfect a representation of company culture as you could find.
That sounds plausible as a rationale. In my experience "bad culture fit" has been used as a stand in for "I don't want this person working here and I don't want to explain why" about 80% of the time.
The reason is often racism. This time it could legitimately be illegal union busting.
Competent HR professionals would not do this to current employees because their documented performance record will render this "soft judgment" more than dubious in any litigation.
There's probably a performance component as well. If you're high performing, can pick whatever job they want, and therefore get paid well above your peers, why would want to join a union?
This was so much more than a job to many of them, just like Bandcamp is more than an e-commerce platform. Bandcamp employed a lot of people who had been there for 5+ years and contributed heavily to its role as cornerstone and defender of independent music. To them, it was as much an extension of their identities as it was a job, and they saw protecting Bandcamp as being equal to protecting independent music. The union came about after they were sold to Epic in an effort to protect not only themselves but also the company and everything it did and represented. Clearly, they were right to be distrustful.
Because unions protect everyone in ways that don't directly relate to pay scale. They help defend against abusive time-off, on-call, or surveillance practices, advocate for pro-worker policies like parental, bereavement, and sick leave, help prevent employees from being fired for using these benefits...
Why do some of the most highly-paid people (actors) join a union and join the picket line? Even if you are high performing, joining up with other workers increases your bargaining position. Why did Steve Jobs join up with Adobe to stop poaching when they were highly sought after work places? Because even if you're a behemoth, you can be stronger in a union.
>Why do some of the most highly-paid people (actors) join a union and join the picket line? Even if you are high performing, joining up with other workers increases your bargaining position.
Perhaps, but there's also a real chance that the union structure you end up with ends up being net negative for top earners. If you're in the top 10%, what makes you think the bottom 90% won't vote for policies that end up redistributing your wages to them?
Also as I said in my other comment[1], whether this is actually true is irrelevant. All that matters is that it sounds plausible and some fraction of people believe it. Perception is reality in this case.
> If you're in the top 10%, what makes you think the bottom 90% won't vote for policies that end up redistributing your wages to them?
We already have the bottom 10% (executives) doing that. We would need evidence to be convinced that your scenario would plausibly happen if workers had more power and executives less.
>Not every talented person is motivated by greed for personal wealth and status.
What I said doesn't require everyone to be "motivated by greed for personal wealth and status". The effect will still be present even if only a fraction of people behave that way.
Of course? The most famously talented people their lives teaching, mentoring and otherwise trying to lift up those around them. Could any person be an effective leader if they didn't enjoy the challenge of working with people with different values and motivations to themselves? I've certainly never seen one.
I will share an anecdote. A friend of mine, his dad owned a construction company. I was talking to him at a party and made a comment that was fairly anti union thinking he’d agree with the sentiment. His response was he loved unions because they gave him access to the best workers. He had 2 unions represented in his company, and his experience was that union workers were objectively better than non union workers. This surprised me because I couldn’t fathom a world where a capitalist would prefer to have unions in his company.
As to why a talented free agent would want to join a union. It seems to me that in an industry with strong union presence then union is obvious to join. It provides so many protections and adds leverage to intangibles that even high earning individuals can’t negotiate for.
Regardless of whether as a high performer, joining a union is actually better for you in the long term, the fact that there's a plausible case against it makes it more likely that pro union employees will be disproportionately affected. Not everyone is 100% bought into unions so I'd expect these factors to play a significant role
No. Unions are for protecting employees against abusive employers.
Your highest performing state is temporary. There will be times where you’ll have off days. You will deal with death in the family. You will be eventually injured. If you aren’t already disabled, you will eventually be (this is just old age). You may become a parent. You might immigrate and come under restrictive visa. You’re a human being with fluctuating states, same as everyone else, and an abusive employer shouldn’t get to power trip over you just because they don’t think it’s legitimate enough for them or something.
So yes, there’s a lot of reasons why a “high performer” might want to be in a union. There’s a lot of life shit we all go through.
Edited to add: this is not the mention your employer might just pull some crap like nepotizing a promotion over you, where a union would come in handy handy!
Is it "abusive" for employers to fire/not pay you if you're away for weeks, or underperforming for months? The terms of the exchange is your time for money. The company isn't a charity.
Even if you think there should be a social safety net for these types of circumstances, it makes little sense for employers to provide it. For one, it has the usual problems of tying important services to employment, similar to how healthcare is in the US. It also puts an undue burden on small businesses. You run a 10 person startup and one of your employees got a long term disability? Congratulations, you have to now find a replacement AND continue paying them. Large companies have law of large numbers on their side, but as an unlucky small business that's 10% of your payroll.
>Edited to add: this is not the mention your employer might just pull some crap like nepotizing a promotion over you, where a union would come in handy handy!
1. has there been a good track record of unions being able to successfully prevent cases like these?
2. Given the level of corruption associated with unions, at least in the US, you're just replacing one problem with another.
> The terms of the exchange is your time for money.
So the contract only covers time? Not actual work, but only time? Do I get to spend the time how I want as long as there is a paper trail that it was your time I just wasted?
> The company isn't a charity.
Yet both are legal and social constructs and not something you can make up on the fly to fit your personal preferences.
> it makes little sense for employers to provide it.
I have been worked to exhaustion for one employer. You don't get to reap the profits and socialize the costs, that only incentivizes more abuse.
> You run a 10 person startup and one of your employees got a long term disability?
So if that person was you would you fire yourself and move onto the street in front of your former business?
This is why we need to both good private (insurance) and public (social benefit) safety net programs in place.
In the case of a small startup, long term disability insurance should cover the living costs of that disability. Yes, that person should be let go, even a founder, if they are unable to perform their duties. But they shouldn't be kicked the street, and the company also shouldn't be on the hook for their care. Either through premiums or taxes, this situation should be accounted for ahead of time. Employment shouldn't be a lifetime obligation of a company.
>So the contract only covers time? Not actual work, but only time? Do I get to spend the time how I want as long as there is a paper trail that it was your time I just wasted?
I'm not sure whether this is supposed to be gotcha at my wording, but it's pretty obvious that if you're paying for someone's time, there's an expectation that they're doing what you want them to do. Otherwise it's like ordering an airbnb but you don't get to use it.
>Yet both are legal and social constructs and not something you can make up on the fly to fit your personal preferences.
Let's go with the legal construct then. Most companies are not in fact "charities" as defined in Internal Revenue Code (26 U.S.C. § 501(c)).
>I have been worked to exhaustion for one employer.
No one is forcing you to work "to exhaustion".
>You don't get to reap the profits and socialize the costs, that only incentivizes more abuse.
If you read my previous comment carefully you'd note that I was only against leaving the responsibility of providing those services to companies. That does not preclude companies paying for those services in some way. Most developed countries don't leave the responsibility of providing healthcare to companies, and instead use a combination of public/private insurance schemes that companies and individuals pay into. Are you against that as well, because that would allow companies to "reap the profits and socialize the costs"?
>So if that person was you would you fire yourself and move onto the street in front of your former business?
In reality there are other considerations for key employees like the CEO which complicates this, but in principle? Yes. The CEO has a fiduciary duty to shareholders and if he's incapacitated and unable to fulfill his duties he should step down rather than using the company as a personal rainy day fund.
> I'm not sure whether this is supposed to be gotcha at my wording, but it's pretty obvious that if you're paying for someone's time, there's an expectation that they're doing what you want them to do.
It was indeed a good question, getting you closer to the truth: No, the employee isn't getting paid for time. No, the employee isn't getting paid for results. The employee is getting paid for fulfilling their terms of an employment contract which may include terms regarding time, results, and benefits that treat the employee as a human being, rather than a cog.
So the main reason to give such benefits to an employee of yours would be because the contract you signed says you have to.
My experience with underperforming workers is that often it is directly the result of poor management. I’ve seen entire teams underperform because of some arbitrary decision from a director. Changing priorities at times when it was guaranteed to remove momentum, or worse, destroy moral. I’ve seen individuals underperform as their manager interrupts them at a concentration destroying cadence. Hell, just 2 well placed meetings can completely ruin a developer’s productivity for an entire day. And then there’s environmental problems. I sat in a cube where the accoustics were such that one particular cube far away sounded like the person was in my cube. I tell you that was hard to ignore. I also once sat somewhere where sales constantly was walking past me. That led to many frustrated hours, and if they didn’t work 2 hours earlier than me, so I could start getting things done at 3pm I don’t know I could have done anything in that job.
So maybe firing someone for underperforming is abusive?
> Unions are for protecting employees against abusive employers.
That's one important role of unions, but it's not the only one. The primary purpose of unions is to allow employees to negotiate with employers on a more equal playing field. Without unions, the power imbalance generally ensures that employees are at a disadvantage when it comes to negotiating a fair deal.
Unions are a protection for both high performers and low performers.
I don't think they should offer protection for non-performers – outside of situations where a life event has mad a performer a non-performer for some reasonable amount of time.
Admission? Like the commenter is some kind of representative?
And while unions aren't homogenous, they generally don't protect incompetence. I've known people in unions that got fired for poor performance, generally. But if I was a cook and cut off a finger while making some company profitable, my performance would certainly suffer while it was healing, and lots of companies world very much rather stop paying me. So in that case yes, I very much hope that a union would protect people from that performance-related loss of employment.
I've known people who managed union employees. While being walked out (fired) for bad performance, they learned the magic words "I have a drug problem". Automatically reverses firing, employee goes to some (company paid rehab) for 1-3 months, gets their job back. They just use this as an option to keep their job, they dont even care. Every "compassionate" benefit you offer will be equally exploited by losers. It seems to me a zero-sum game.
The exceptional number of people I've known in Unions for decades are mostly career professionals with good professional ethics. Your anecdata vs mine. I hear garbage like that from anti-union people but have never seen that in reality. Is it a union for bank robbers?
> And while unions aren't homogenous, they generally don't protect incompetence
They tend to insist on due process for adverse actions, and management tend to hold out the idea that being able to dismiss arbitrarily without evidence or process is essential to efficiently dealing with incompetence.
Your assertion tends to rely on your anecdata which tends to not be any more useful than anybody else's. I've known a huge number of people in various unions and this simply does not reflect any reality I've experienced in 25 years of work experience.
Some of the most important things unions negotiate for aren't about pay and benefits. They negotiate things like working conditions, handle disputes, and the like.
Those things are usually much more important than pay and benefits, and when union negotiations stall, it's more usually about that sort of thing than about money.
I hate to bring up the old example, but it's used because it nicely illustrates the value of unions for everybody -- not just the union members. Things like having a 40 hour workweek and weekends only happened because of unions.
False. H1B visas given to work for a university are "cap exempt." They allow to work for universities and other nonprofits only. H1B visas to work for a for-profit company can only be obtained through a lottery.
There's more nuance here. There're ways to "port" your H1B to another company (incl. while working part-time with one leg at a non-profit and another at a startup), but few want to go that route because of all the other bullshit we already had to go through.
I don't think we are in disagreement here. My main point is that many feel "pressure" from their employer. Knowing and keeping your options open and ready to act on them, even at other universities, empower all workers.
If I'm reading the plot from the article right, the most impressive gains have happened in the first half of 2019. Since then, the progress has been notably slower, with year-on-year performance increases for some network architectures falling behind Moore's law already.
