NaibofTabr

#566

“Hey, listen!”

The logo, the UI, or the code?

Arch often seems to ignore the fundamental rule:

don’t break userspace

Linus is in the right. Arch developers are frequently in the wrong.

Then a visit with your doctor, to discuss cholesterol levels and hardened arteries.

battered & deep fried

If you’re just doing a quick config edit, nano is significantly easier to use and is also present in most distros.

Vi/Vim is useful as a customizable dev environment, but in the present there are better, more feature-rich development tools - unless you are specifically doing a lot of development in a GUI-free system, for some reason.

The Tragedy of systemd

Oh no, it’s worse than that… we use the metric system to measure the customary system…

The Mendenhall Order marked a decision to change the fundamental standards of length and mass of the United States from the customary standards based on those of England to metric standards. It was issued on April 5, 1893, by Thomas Corwin Mendenhall.
[…]
Mendenhall ordered that the standards used for the most accurate length and mass comparison change from certain yard and pound objects to certain meter and kilogram objects, but did not require anyone outside of the Office of Weights and Measures to change from the customary units to the metric system.

https://en.wikipedia.org/wiki/Mendenhall_Order

Technically every unit in the US customary measurement system is just a weird conversion factor of an equivalent metric unit. At this point 1 yard was defined as 3600/3937 meter, which means 1 inch = 2.54000508 cm. By 1959 everyone finally agreed that this was stupid and redefined it as 1 yard = 0.9144 m (1 inch = 2.54 cm).

All measurements in the US are based on standard reference objects provided by BIPM.

naturally

Oh, so all those examples about AI destroying the world by making paperclips… this is what they were talking about!

'O brave new world, that has such people in ‘t!’

By the phone company.

Heh, that won’t stop a C-level from thinking that you just write code.

The nuke plant is, of course, Historic Chernobyl No. 4, now a war memorial.

For individual projects the way this usually works is one of the larger companies that rely on the project hires the developer as an employee to maintain the codebase full-time and help integrate it with their internal processes.

Larger projects might form their own company and sell integration & support to other companies (e.g. Red Hat, Bitwarden).

Otherwise you’re basically dependent on donations or government grants.

There’s a Wikipedia article on this subject: Business models for open-source software

And there’s various industry opinions:

Demystifying the Open Source Business Model: A Comprehensive Explanation

How to build a successful business model around open source software

Open Source Business Models (UNICEF course)

I think monetization is easier for user-facing software though, which a lot of this material is written around, and harder for projects like libraries.

It is an absolute PITA to keep an email server on the “nice” list so your company’s email traffic doesn’t get spam filtered by every service provider, and the major services (gmail, outlook, etc) are all federating their spam filter lists so many times if you get blocked on one you get blocked on all. There is so much spam to deal with that the filtering is highly automated and there’s little human oversight.

The point being, it could only take a handful of incidents reporting a company’s email as spam to ruin their reputation and result in email from their domain getting automatically filtered everywhere. So, you know, if they don’t support an easy way to unsubscribe then they are in fact behaving like spammers, so flag them and let them deal with having their domain blacklisted.

Fantastic. Hadn’t heard it before.

And it only covers a period of 51 years (1948-1989).