Choosing a tech stack for my card game platform¶

Here are some of the technologies I have professional experience with, in no particular order.
Let's check their pros and cons for this card game platform project I want to start building...

Note that these bullets points are strictly subjective and personal - I think that all these stacks are perfectly good and valid, and choosing one or another is always a good option anyhow!

Next.js¶

Pros
- Modern, evolving at a quick pace, tons of best practices built-in. Very well documented.
- I love TypeScript, I like React
- Now that it's rather mainstream there is a huge ecosystem for it
- One language (TypeScript) to rule them all!
- With new runtime such as Cloudflare Workers, Deno and Bun coming in, and all converging towards the use of standard Web APIs, it's a quite exciting time for JavaScript on the backend!
Cons
- Even though it can be used a fullstack framework (as Theo, from ping.gg fame, explains here), and despite the numerous benefits of Node.js...
  I can't help thinking that Node.js doesn't give me the same level of productivity than what I can have by using a "fully featured" mature backend framework such as Rails, Laravel or Django - where all the core features such as a typical backend, like an ORM, database migrations, logging, etc, are features that are already plugged together out of the box
- I gave a quick shot at Prisma, the trendy Node.js ORM; even though it has some undeniable qualities I didn't really fall in love with it - for various reasons.

Laravel¶

Pros
- Probably the framework that comes with the highest number of features built-in.
  Asynchronous jobs, websockets, authentication, support of Vite.js, you name it... Laravel has it by default.
- Easy deployment: whether it's via Forge or Vapor, Laravel comes with charged-but-very-handy solutions for deployment.
Cons
- PHP has been my main programming language for backend stuff for more than 15 years, and even though it's never been better than today I grew a bit tired of its $peculiarities

Ruby On Rails¶

Pros
- Almost as "all features built-in" as Laravel
- Huge and mature ecosystem
- Ruby is a very expressive language
Cons
- After having worked with languages that have inlined type annotations during the last few years, I struggle with languages like Ruby which don't have this
- So much magic that it can be really hard sometimes to trace what method comes from where and really understand what's going on

Django¶

Pros
- Python
- My favourite ORM, with (for me) the right balance between power and pragmatism.
  Also comes with great support for modern database features, such as JSON operators.
- It's been my tech stack for the last 4 years, so I'm pretty productive with it
- Has excellent GraphQL libraries (and I love GraphQL! ^_^)
- The Django Admin website is a huge gain of time while prototyping stuff
Cons
- Some old-school aspects
  (no routing via HTTP verbs for example - which can be solved by using Django REST Framework, but for some reason I've never really liked it )
- Features such as websockets and asynchronous jobs are not built-in, but Django's ecosystem is vast enough to be able to add them with a good integrations (Channels for the former and Dramatiq for the latter)

Go¶

Pros
- I love the minimalism of the language
- Very stable over time
- Strongly typed, with a very smart compiler
- It's really nice to work with a programming language that have features such as code formatting or testing built-in
Cons
- Whether it's at the database or the GraphQL layer, null values have to be handled and I'm not a big fans of the solutions Go has to offer for that (sql.NullString and friends, or pointers to primitive values)
- Can be really verbose sometimes