ocius

Regarding the first point: owners typically pay themselves a salary in addition to owning the company and thus benefit from both. Employees do not have this advantage.

The two problems are not mutually exclusive, and certainly unskilled labour is a significant problem for a good fraction of the population; but if this were the only problem, then we wouldn't see the amount of wealth concentration and decoupling of productivity from income that we observe.

The public already gets its share via taxes, I'd argue, the problem that I see right now is really that the wins for the employees are (usually) limited to their salary, whereas owners get everything that's left over. And the more you already have, the more you are able to gain..

Most such discussion suffer from an inadequate definition of what is meant by "socialism" vs what is meant by "capitalism". The most important aspect to me is ownership / control of the means of production, and I think there are very interesting compromises to be had in between extremes. What I mean: imagine if it was normal for companies to pay out a large part of their earnings to their employees. This does exist, of course, in some of the best companies, but is not widely spread and even when in place, the share of profit going to the employees is too small. Another share could still go to shareholders that "didn't work" for the profit to incentivize investments. Seems like a small change that could have a huge impact on wealth distribution.

I have anecdotal evidence that at least a part of this is due to Musk's behaviour. My father is in charge of the car fleet for a large German company, and they have decided to exclude Tesla from their fleet because of Musk's recent behaviour. At the same time, they have also decided to go full electric for the entire fleet (no hybrid either).

Installments of this type are not very helpful, as they proliferate the notion that the so called "debt" of a nation in its own currency has to be repaid or is in any way bad. The goal of a nation with a sovereign currency should be to keep prices stable while advancing wellbeing in general. Often, this means spending new money, which is framed as debt and said to increase inflation. In reality, new "debt" can both increase or decrease inflation, depending on how the money is spent. There is no direct correlation between increasing the pool of money and inflation. If money is spent on a product that currently has low demand due to a crisis, prices won't be increased and there won't be inflation. If money is spent on a product with high demand, then prices will increase and inflation will worsen. Example: after COVID, construction was down 30% in Germany. Government could have stepped in and spent new money to use the available labour, but instead it remained unused. This way, the labour went to waste, and construction companies, like many other companies, had to charge more for the same services because there was less demand in total, which causes inflation.

Just as a side note because you mentioned it in relation to DOGE: the US government cannot go bankrupt in its own currency. As long as prices are kept in check, there is no reason not to create more money. The fact that the US has to pay interest on it doesn't change that. If the government decides to spend more money than it takes in, the money is simply created by the central bank (through the detour of bonds). If you are interested, read up on Keynesian economics and Modern Monetary Theory.

We're expecting the amount of data to double every year for the next couple of years. That would require a lot of storage hardware and maintenance cost, don't you think?

We are expecting the amount of data to at least double every year for the next couple of years. That would mean that in 5 years, we'd have to manage more than 500 16 TiB hard drives in our office :D

We don't really have that many constraints, despite that it should be cost efficient and save us from loosing all our data.

Some good points! There are, however, also changes that you do not want to propagate to your teammates or production unless you are 100% sure that they are the final solution to the problem you are trying to solve. Sometimes, the complexity of a problem is so high that you have to experiment with the code.

When those changes include non-reversible database migrations, for example, it is better not to integrate that code into "mainline" before you've settled on a specific implementation.

Good point. You are not dependent on Cloudflare, though, you can use other similar services or open the port on your router if it supports it. And if you pay for it, you can get an SLA so that they can't mess with you by terminating your project.

I would still consider it very nice of them to offer this free service that let's you break free from your dynamic IP, crappy router and at the same time giving you protection that you couldn't set up yourself.

I think many people just want to be able to make their device available from the internet - this type of liberty is not really important for many people.

I can recommend the Rasperry Pi for self-hosting, if you pick one of the more recent models with more RAM (3 or 4 with 4GB of RAM at least). You just need to keep in mind that the availability of your website is not quite as good as if you were hosting in a datacenter.

Regarding security: you might want to have a look at cloudflared. This is a (free for small projects) service by Cloudflare where your server (Raspberry Pi) connects to cloudflare, and all HTTP traffic is proxied through Cloudflare. This has multiple advantages:

- you don't need to open a port on your router that forwards to your Pi, which is good for security and simplicity (some routers don't even have this option)

- you get all the cloudflare protection with the click of a button, including HTTPS with automatic Cloudflare certificates, DDoS protection, etc

- you don't need DynDNS to point to your dynamically changing IP

Python is simpler than Java. This might not be a very important factor in software engineering of large products, but it is very important for prototyping.

Also, a lot of research in ML is done in Python, and there are quite a lot of startups doing ML, so Python is a natural choice. Same for data science.