Hacker News new | ask | show | jobs
by miiiiiike 448 days ago
I saw the talk last year. Let's do it.

If I had $100,000 to spare I would give it to Django as an unrestricted donation. It would be worth every penny. But I'm a solo-dev and don't have that kind of cash on me.

Here's what I can do. I will send you $1,000 today if you can get something like a `django.contrib.rest` package (i.e. official rest api support) on the roadmap and secure matching funds. I'll make it a recurring donation if development gets going.

I don't want to support other REST libraries because there's no consensus. One app ships DRF views, another Ninja, and that one over there still has a Tastypie dep. Remember Piston? Using multiple rest packages means that figuring out how to create a consistent error format, etc is an exercise left to the reader.

Django makes great design decisions that the majority of devs can get behind. I'd love to have an officially supported REST package that feels like the rest of Django.

`django-tasks` is a good recent example. I was reading through the code a few months ago and I was impressed.

1 comments

Another project I'd fund: Making Django an async-first (not async-only) framework over the next decade. It's easier to mix sync code into async than the other way around.
If one can make it without breaking the API too much, why not.

The #1 reason I use Django is stability. I have projects that span 4 major Django version without any significant break. That's quite a feature.

So glad that's not actually true so I didn't have to rewrite sqlalchemy from scratch
It's more straightforward to call a sync function from an async function than the other way around. That's a pretty uncontroversial statement. Adding async support to a library doesn't mean starting starting over.

I'm not fully understanding your thought. Can you say more?