I think you hit the nail on the head.

Elected officials unfortunately don't have that much incentive to hire the "cheapest" company as the debt will be incurred over the next 100 years while they will be long gone.

They probably hire the company that they feel will give them the least amount of trouble, which is the easiest to navigate or that will do something for them in exchange. It's the "not my money" issue at play.

I think it's quite the opposite. There is heavy incentive to pick the lowest bidder without any consideration paid to how absurd the bid is, which is where you get these crazy cost overruns. The bid was never doable to begin with.

And once there is a cost overrun it's easy to bloat the budget because the government is stuck with the contractor they hired.

It makes sense at scale. If you are a company of two there are probably better solutions.

At scale, you can very granularly define policies for each secret. When a secret is accessed, it is done so through a user or application identity. Each access is also logged.

So then how do you manage the secret that authenticates an application's identity? And what good is the logging if after an application has the secret it can do whatever it wants with it?

if it is an instance on the cloud, GCP and AWS let you define ServiceAccounts that get populated on the Instance at boot time.

you should only let the instance access the secret it requires.

and how do you manage secrets that let you define that ServiceAccounts?

As OP wrote, you did not solve it, just moved it to a different level.

If you use Vault, you should use it as an RBAC system as well.

That means that each application got a ServiceAccount (SA) and each user got a username/password. Based on your identity, you get access to specific secrets from Vault.

I'm using Kubernetes extensively in my day to day work and once you get it up and running and learn the different abstraction, it becomes a single API to manage your containers, storage and network ingress needs. Making it easy to take a container and getting it up and running in the cloud with an IP address and a DNS configured in a couple API calls (or defined as YAMLs).

That being said, I will also be the first one to recognize that PLENTY of workloads are not made to run on Kubernetes. Sometimes it is way more efficient to spawn an EC2/GCE instance and run a single docker container on it. It really depends on your use-case.

If I had to run a relatively simple app in prod I would never use Kubernetes to start with. Kubernetes starts to pay itself off once you have a critical mass of services on it.

One could argue if you have a tiny set of services you are better off using a managed offering like AWS Lambda or Cloud Run

There are organisations with 1000's of services on Serverless seeing enormous benefits in reduced management overhead and reduced costs compared to the Kubernetes solution they previously ran.

My issue with serverless though is that you need to refactor your code to make it work specifically for it. If you don't start to think serverless on day one it gets more and more difficult to convert to it down the road.

It feels like this became the life-cycle of Internet.

The same thing happened with decentralized websites and blogs.then everyone got attracted by the managed platforms and now the web is more centralized than ever.

It pre-dates the internet. Happened with cable TV as well.

you make the right point. A lot of people see this as youtube censorship, but the podcast was also available in its purest form: through a RSS decentralized feed.

Now it moves to a walled garden with content unavailable to the outside world.

I used to think this, until:

1. We had a child

2. We had multiple children

3. We realized getting children (who may be sleeping) into a car for even a 10minute drive becomes a big production

4. We got rid of our car

Also, other reasons:

A. It is 8:30pm, you're at the office, have another 3hrs of work to do and cant spare even 10min to get away. Very common in my Junior Analyst days. In fact, we had a company sponsored SeamlessWeb account that we could use anytime.

B. You are on a business trip at a random city/hotel w/o a car

C. Your car is in street parking and you dont want to lose the spot (wicked, i know...)

> Did everyone really become THAT lazy that driving 10 minutes to get your meal is that much trouble?

Depends on density, and traffic.

Getting to a nearby restaurant to pick up dinner, even a close by one, would easily take 30 minutes+ round trip. If I want food from someplace more than a couple miles away, make that 45 minutes or more round trip for dinner.

Or I can order from an app and have food delivered.

The question then becomes, is saving almost an hour of time worth $20?

Meal delivery doesn't just exist because people are lazy. It has been around a long time via the much more inefficient process of calling a restaurant that you already knew about, having someone spend time with you on the phone getting your order and credit card # and then dispatching a delivery person they employed directly to deliver the order to you.

My experience is that calling for pizza is faster and easier than using an app.

I'm in Toronto, so we have Skip (Skipthedishes) instead of GrubHub along with UberEats and Doordash

1.) Skip doesn't allow the restaurant to jack up the price, so to the customer the total cost is the same 2.) These companies toss out tons of coupon codes and referral codes that bring the overall cost down (sometimes even cheaper than ordering directly from the restaurant) 3.) In dense urban centers, a ten minute "drive" is way more challenging/time consuming/effort than it would be somewhere else. In fact, these services use bicycle couriers in these areas.

I've probably only used each of the major delivery apps once or twice, so I'm not representative of their customer base, but yeah, every now and then I'm having a specific day where I'm feeling that lazy (and of course driving anywhere in the bay area around dinner time is likely to take a lot longer than 10 minutes round trip). Then again I'm living in the bay area and not making anywhere near FAANG money. I can definitely see the delivery fees being negligible compared to the value of my time if I was making 2-3x my current salary.

For me, it's sometimes laziness, but usually not. The long and short of it is that delivery only happens when it's hard to leave the house for whatever reason. If getting myself to the restaurant is an easy option, then dining in generally is, too. Takeout only happens when I've been tasked with picking up burritos on the way home from work.

Not everyone has a car.

It's the opposite. People are so busy that they can't spend 10 minutes to get food.

Also the fact that they can afford the premiums in the first place implies that they're not lazy ;)

I don't get those delivery companies (Doordash, Grubhub, etc).

Even with a first-timer promotion, a meal would end up being close to 25% more expensive than ordering directly with the restaurant.

I'm happy to order directly with a local restaurant on the phone, drive 10 minutes (especially now with no traffic) and get my food on my terms. I end up paying way less and the restaurant keeps the full dollar amount. It's a no-brainer win-win situation. I wish more people would order directly with restaurants.

Some fastfoods, with quick preparations and basic food can actually benefit from services like DoorDash. If they don't like technology, are not marketing specialists, live mostly offline, then it can bring them many new customers. So I think it's up to the restaurant to decide. Premium restaurants may have money and skills to have their website with their delivery ordering system and returning customers. It varies, I guess...

This must be one of the most misunderstood pieces of popular science.

The Coriolis effect is WAY too weak to affect the direction the water is flowing. Typically the way the water moves initially and the shape of the container will have an outsized impact in the final direction.

I think the issue is between people who interpret the statement as "water is predisposed to drain in a particular direction depending on hemisphere" or as "water will drain in a particular direction depending on hemisphere".

Does it have an effect and can it sometimes be seen? Yes. Is it often swamped by other factors and rendered irrelevant? Also yes. That doesn't mean it's false, but it does mean that some statements that rely on it as a factual basis are at best mistaken.

When you have a large reservoir with a drain, keep it still, and the water reaches the level where it starts to rotate, the Coriolis effect is the dominant influence.

It is something large enough to be plainly visible, but not something that appear on every drain.

If you are an engineer those type of stories should make you rethink your usage of Google Chrome. Chrome having so many users empower them to implement those type of nonsensical policies.

As said in other comments it is trivially easy to switch to Firefox (or any other browser you feel that fits your needs better).