To explain

There are 2 Error struct / enum declarations, probably in separate files

To the ?, they are different types and cannot be converted from one to the other (because they are two disparate structs that happen to have the same name, but can have different bodies)

To fix this

You can either use .some_func_result().map_err(|err| /* conversion here/*)?; +

Or you can impl From<Error1> for Error2

And you should also name it ThingError, so you can visually differentiate the two

+ There are like 10 different mapping functions, depending on if you’re using an option or a result

I never remember which one specifically i need, (unwrap_or, map_or, map_or_else, ok, ok_or)

I usually just hunt through the auto complete list until i find the function signature that gives me what i need

All distros are equivalent, as far as software is concerned. They all have access to the same open source software, and Flatpak; AppImage; and Snap can be used for extra portability.

Think of a distro like a pre-configured image of linux. You can always change the configuration later, if you desire. For example, the Desktop Environment. All you have to do is just install a different DE package (usually via command line)

The DE has a major impact on user experience. Use KDE plasma for a more windows-familiar experience, or Gnome for a more Mac-familiar experience. Or experiment with others

The Linux Experiment is a good resource

Linux is already at 4.5%

https://gs.statcounter.com/os-market-share/desktop/worldwide/

It haunts me in my dreams

We kinda do, with GPS satellites that have to correct their clocks due to the effects of gravity and speed

And communication with space probes

I use 24h clocks and ISO 8601 dates almost always

Honestly, I’m better at organizing code than I am my actual life

Timezones make intuitive sense for humans

UTC / Unix timestamps make intuitive sense for computers

The issue is bridging the gap

It’s an explicit “opt-out” by the OP, such that their content cannot (legally) be used to train LLMs or such (Chat GPT, Github Copilot, etc)

Well, that’s what I assumed until i read the license terms. It doesn’t explicitly mention AI or LLMs, but it does say

You may not use the material for commercial purposes

Which i assume has the same limitations for AI training, for commercial AI

(I am not a lawyer)

While i also disagree with python’s tendency to use exceptions as control flow

Python is a pretty stellar scripting language. I wouldn’t use it for app dev, but it’s quite handy for the odd automation or CLI task

I feel like that would cause false positives

Then again, not that many people mess with their router

It’s the same for me

Except, i try to give reddit as little traffic as possible, unless i need it for reference for something