danjac

"Always hiring" could mean rapid growth: for 2-3 years though that may mean a lot of chaos as you are constantly square-bashing new hires into teams.

On the other hand, it could mean high turnover. Looking at their Glassdoor reviews and filtering out their HR-generated fakes, it paints a picture of both scenarios: https://www.glassdoor.com/Reviews/Flexport-Reviews-E1011163....

Agencies do not exist to build great software. I mean sure, it helps if their output is good enough to guarantee more work, but that's not why they exist. Their purpose as a business is to generate billable hours.

Customers do not hire agencies with the expectation they will build great software. They turn to agencies because they don't have the time or skills to do it themselves, and agencies are cheaper than hiring people and they are more dependable than freelancers. By that I don't mean that freelancers are collectively unreliable, but with an agency you have a dozen or more developers and if one falls sick, they can just replace them with another developer. If your product depends on that one freelancer, you are screwed.

So "10x developers" aren't really suited for agencies, where every hour is billable and accounted for, and the output does not have to be great, it just needs to be more or less reliable and more or less on time and under budget (or at least cost overruns are accounted for).

It's not happening fast enough. For greenfield projects, new startups, sure.

But the rest of us stuck in legacy-land the toxic waste dumps of these over-engineered behemoths will be with us for years to come.

He's a made man, as the Mafia used to say. Once you're in the circle you can keep failing upwards.

Where else are the investors investing?

In the 90s complete amateurs were able to make "visually interesting results" with little more than Notepad and a copy of "HTML for Dummies".

Any artist or designer has to learn their medium before they are allowed to produce designs for it. If you want to learn to sculpt, you have to learn what you can and cannot do with clay; if you want to learn to design clothes, you have to learn about fabrics, sewing and tailoring (Jean Paul Gautier and Versace worked as tailors in the industry while they were doing their early work).

Yet there seems to be a free pass where a "web designer" is allowed to dictate how a site should work and look without knowing the basics of HTML, CSS and JS.

If you are wandering the desert, dying of thirst, and you find a bottle of Coke, you should probably drink it and not worry too much about getting diabetes.

They don't want it thanks to decades-long misinformation campaign waged with fossil fuel money.

The Finnish Greens are pro-nuclear, perhaps because they are not beholden to the same sources of funding as their German counterparts (and not to just blame the Greens here: look at your former Chancellor Schroeder).

Sigh. Oligarch money still flowing, then.

All true, but right now those risks pale in comparison to the ongoing disasters caused by fossil fuels.

Above all, journalists should be skeptical of Nate Silver.

The German Greens are slowly inching towards nuclear, now the original 1980s membership are dying off and the Gazprom funding is running out:

https://www.dw.com/en/german-greens-lay-out-nuclear-power-po...

It's not too bad. At least, not nearly as bad as naming your language after the most common English verb of motion, especially when your company is literally Google itself.

Not sure what fly.io offers vs Heroku or others (I have played with it some time ago but not used for anything serious), but for an equivalent I'd be looking for automated load balancer setup with SSL, easy scaling up so I can go from 1 to 2 or however many web services (with UI or CLI), simple deployment configuration with a Procfile (or whatever) and managed PostgreSQL/MySQL/Redis including backup/restore when needed.

That's more than what I would have or need locally.

There's a sweet spot of early startup or side project where you don't have the time, budget or people to manually set up and maintain servers on your own or deal with the complexity and cost of Kubernetes or AWS, especially when your focus is on building the product and acquiring customers.

Heroku (before its inevitable enshittification under Salesforce) was great for this use case. Sure you will outgrow it at some point, and it did get expensive, but when you just want to throw up an MVP with minimum fuss and maintenance you could do much worse.

I use Dokku on top of Hetzner for my hobby projects - hosting is super cheap, for a little extra I can add a mounted volume for storage, and if the project outgrows a single server I can always just break out of Dokku and use some Docker containers behind a load balancer.

If you are outside of Europe, Digital Ocean or Linode may work better for you.

For hobby projects - where I dare not touch AWS for fear of going bankrupt from a misconfigured service - I found the sweet spot to be Dokku on top of a Hetzner or Digital Ocean instance. It provides a Heroku like interface on top of cheap hosting, and is fine where you don't expect to scale very much.

> It's historically very abnormal for adults to be so fascinated with dragons and other escapist features.

Literally the oldest surviving epic is Gilgamesh, which features Tiamat, the mother of dragons.

Also, what happened to proper lighting? Even in a night scene, we would actually like to see what's going on, otherwise I might as well listen to a podcast drama.

Given actual vs perceived social mobility in the US compared to Europe, it could be argued that the US reached that point some time ago:

https://www.economist.com/graphic-detail/2018/02/14/american...

I quote:

"IDEs are easy mode and waste of resources. Its ok to start with them as a noob, but down the line i want to customize every part of my toolchain. (And thats just horrible to do in VSCode etc..)"

I agree that vim is a perfectly fine text editor.

Yet half the (neo)vim people in this thread are yelling it's the One True IDE that can, in fact, do everything, and if you are using VSCode or IntelliJ or whatever then you must be a noob or too lazy to invest time in your tools.

> Again, I would like to stress, I am not trying to make any objective claims as to how correct your approach is. I am simply observing that there seem to be broadly two categories of programmers. Those who feel like investing time in learning and writing/modifying "tooling" is worthwhile and those who do not. I feel like you fall into the latter category.

I use zsh, with customizations. I know git, and mercurial, and a few other VCS I've forgotten. I know my way around the Linux toolkit. I can configure a Python or frontend or Go environment with whatever linters and formatters and pre-commit and so on and so on. I have a fat repo of dotfiles. Hell, I use vim whenever I need to work on headless servers or the odd config file and the like. Works fine, in that use case.

Every time I learn or modify a tool, I look at whether there is a net benefit in learning it, whether it makes me more productive. Zsh shortcuts for example are tiny, but I get a nice little productivity boost that adds up to many hours over time.

I have spent the hours learning and configuring neovim, but run up against that point where that time spent is no longer yielding that benefit, compared to just using vscode and putting up with its issues.

So, I chose one tool over another. But it's a generalization to say I somehow don't think learning "tooling" is worthwhile.

Yes I did, and no it didn't really help. Sure it was a much slicker starting point, but when I needed to do a bit more (ie. configure a formatter) it just added more obfuscation than anything. The "pre-cooked" part does help with setting up linting, but formatting...you're kind of on your own (if you search the source you can find an example with prettier but you will still need to add your own integration for anything else).

> I agree that neovim is not for people who are not interested in learning their tools.

That, right there, is one of the biggest turnoffs I have encountered from the community. The sheer disdain in that sentence alone.

I learn many tools, and am deeply interested in learning them. That includes languages, frameworks, databases, infrastructure tools, operating systems, standards and protocols. I wouldn't be in this job otherwise.

However, my time is limited. I have to be selective in what I spend time learning. Much of that choice is determined by my employer, project and other requirements. Free time left over for learning other things is even more limited.

Learning another language and API and plugin ecosystem just so I can configure my editor? Maybe for some people, that's time well spent, but personally - having spent a fair bit of time with neovim to give it a fair shake - I realized that what I don't need is a perfect artisinal text editor, but a good-enough IDE that lets me do the things I'm either actually interested in, or paid to do.

Vim as a simple text editor? Perfect, use it every day when dropping into a headless server to edit a config file or two. For an IDE? I don't love vscode, but for what I need, it's fine.

I don't need an IDE for Python. Of course I can use a plain text editor if I want. Hell, nano even.

But right now, today, at work, I have a large legacy Django project to maintain, and it's kind of nice having little things like auto code formatting and linting to help me as I go and prevent me from making dumb mistakes and saving me a ton of time so I can focus on the problem at hand, not making the code look nice for review and I can hop between buffers to fix up the tests as I change the code.

I can certainly just run black/isort/ruff what have you manually, or with pre-commit, sure. But it's more productive to be able to see changes as I work.

Sure maybe I shouldn't be using Python, and maybe the original developers should have written it in Rust or whatever. Fine. But that's not what I'm being paid for.

I use vanilla vim when I have to ssh into a server and work on some config files. As such, I much prefer it to nano, and it's installed, or easily installable, on any Linux server.

For that use case, it's absolutely great. No complaints.

My feeling is that neovim per se, as a piece of software, is fine, but that it belongs to one of those developer communities that prizes freedom and infinite configurability over the day-to-day practicality of getting things done. So something like LSP integration has too many ways to do it across too many plugins.

If you are deeply involved in the community and are willing to put in the time, sure, you could probably build something superior to vscode. But the time invested to get there might not be worth the marginal productivity gains.

The problem is when you are a newbie to a space, you don't know the difference between "cool" and "what is the best practice as defined by the experts".