Most of these things can be packaged by distros up to the desert island standards, albeit with great effort.

So it can be done.

It's just that people who write that software, and the tooling, rarely care. Which makes things harder for everyone else. Harder than it could be.

t. spent 8h on something that should've been 2 commands

4/4

But also, it's very accessible:
apt-get source -> apt-get builddeps -> make changes -> dpkg-buildpackage -> dpkg -i your-freshly-built.deb

I'm sure many other distros provide a similar solution. Some default to downloading sources as part of building the package, but AFAIK most still require that the actual build doesn't rely on internet.

Can you say the same about your average docker image? Flatpak? Android app? Rust crate? Golang module?

3/

I remember my dad's RedHat came on 4 CDs: 2 Binary Discs, one Source Disc and one Documentation Disc.

If it worked the way I think, you could take it to a desert island and rebuild everything from source, with just those 4 CDs and an x86 computer.

Today a Debian mirror has 100+ GB of source packages and 600+ GB of binaries just for amd64.

But at least it's one place, heavily replicated. And IIRC it's supposed to work on a desert island.

2/

I've said this before but

Software freedom is not when the source code for the program you use is somewhere on github, has 5000 dependencies and doesn't build outside of CI

Software freedom is when you have the source on your computer, modify it, build it, and run a modified version

How much of the software you use on a daily basis would you be able to build from source? How many urls does that require fetching, and how many of these can stop working any time without prior notice?
1/

The year is 2023. Most people in developed countries have broadband access to the internet. Long gone are the days of 56k modems and waiting for pictures on a website to load. These days we wait for the text to load.

Putting emoji in unicode was a mistake.

I wonder why nobody I know of uses that notation...

I was wondering if there's a concise way to write down communication protocols.
Something that'd be to RFCs what Relation Algebra is to SQL.

And then lo and behold, there exists a thing called "Algebraic Specification of Communication Protocols" on cambridge website.

It's a book.

You have to pay for it if you want to read it.

ffs....

People say this place is a street.
But they behave like it's their living room.
While in reality, it's a soapbox on the main square.

Meanwhile all I wanted is a watercooler.

Already said that, but that was on niu, and I want it pinned here, so...

Theory:
1. Reinvent a square wheel
2. Understand what's so hard about making wheels
3. Observe how a round wheel someone invented a while ago performs better than your square one.
4. Appreciate the brilliant and simple solutions to all wheel design issues that a round wheel implements.
5. Throw away your square wheel.

Practice:
1. Reinvent a square wheel
2. Push it to production

I've already posted this but it was on niu and I need it pinned on my profile, so....