Monorepos - How the Pros Scale Huge Software Projects // Turborepo vs Nx

The problem with these Monorepo tools is that they assume that the code is only written in Javascript ecosystem. This is not usually the case. That's definitely not the case for Google's internal codebase where they have different systems that communicate with each other using Protocol Buffers.
Tools like Turborepo and Nx have definitely changed things, but they are still a lot limited.

JS developers discover cmake

Cool - like Gradle!
Except that build caching sharing - that sounds like fun

lol so many "microservices" make the mistake of checking the source into a mono repo

Sadly, these options are not available for Python, are there any alternatives for Python?

Hi, great content!!!

Do you have any video where you explain the differences between monorepo and microfrontend? and if it is possible to use both?

I think you should mention an update that now Turborepo is now written in Rust not GO anymore

Webtatds reinventing make

Damn 1 year ago you didn’t even yet have 1 million subs 😮

Was the speaker in this video an AI? Why does the voice sound so robotic?

Let's be clear, you as an individual viewer do not need a mono repo. If you work for a medium to large scale organization, perhaps your organization could benefit from a mono repo. But you as a dude who writes code for his home lab in his underwear do Not need a mono repo.

bloat lets gooooooooooooooooooooooooooooooooooooooooooooooooo

I love the fact that you can download the cache. Oxymoron in action.

As long as you don't put totally unrelated stuff in a single repository it's totally fine to have a single repo.

As soon as that is not the case anymore (think of a completely separate product) use a different repo.

But really don't split up a single app into multiple repo's (like backend, frontend, serverless functions etc). Now to run the app you have to checkout multiple repo's... This will bring you hell. Don't get me wrong you should have proper isolation between code but a separate repo is the wrong type of isolation.

If you split it up then it should be in such a state that a separate team can work on it with their own release cycle without bothering you.

I like hockey

Companies have to be careful to not think it’s ok to store they much code in Git. They see Google do it then think it should be done in git. It’s typically not a great idea to store that much in git.

The voice sounds different.... hummm

I would love to pull that repo