In 2019, which is the best stack to start a web development project?

I think these are the major ones right now (in the open source or non-enterprise world):

Back-end Stack

1. PHP
   • Symfony
   • Laravel
   • CodeIgniter
   • CakePHP
2. Python
   • Django
   • Flask
   • Pylon
3. Node
   • Express
   • Hapi

---

Front-end stack

1. SPA
   • react
   • angular (1/2)
   • vue.js
   • backbone.js
2. Non-SPA
   • jquery
   • d3.js
3. User Interface
   • Bootstrap
   • TailwindCSS
   • material-ui
   • angular-material
   • primeng

---

Which among these (back-end + front-end) do you think has the best chance of future-proofing?

I haven't included Java and ASP.NET because they're only used by large enterprises and their market share is minimal in web development.

Then there are some niche back-end stacks too like golang, ruby and erlang but either their share is minuscule or on the decline.

Comments

[Andrew Truex]

Vue.js has the least steep learning curve for frontend frameworks and has been gaining considerable momentum lately. I typically use a node.js backend to run my web applications. d3.js is only necessary for complex data visualizations. As for UI, it comes down to personal preference but if I had to say one is "future proof" it'd be Bootstrap. It's a simple and straightforward introduction that beginners will continue to utilize.

[Prahlad Yeri]

Have you tried TailwindCSS? It seems to be gaining more traction than bootstrap these days.

[Javier Aguirre]

Regarding User Interface, I think Design Systems are the way to go for many companies, github.com/alexpate/awesome-design...,

We’re using one as base for creating our own, they usually have React/Angular/Vue components.

[rhymes]

I think it really depends on what you have to do.

I'd start with a serverless platform in 2019 for starters. You can find serverless runtimes for many of the platforms you mentioned.

Be it serverless or not, I'd say that any is valid, the one you know better :-)

I haven't included Java and ASP.NET because they're only used by large enterprises and their market share is minimal in web development.
Then there are some niche back-end stacks too like golang, ruby and erlang but either their share is minuscule or on the decline.

I'm not sure this is true at all, it's not like there's an easy way to measure a share of a language.

Is it number of web servers running a language? Is it the traffic these web servers are handling? Because both Go and Ruby don't come out as "niche" to me.

Same goes for Java and ASP.NET, it's really hard to say with confidence "hey don't matter". Based on what? :-)