Getting started with an react/apollo app is not simple. We could use meteor, but what if you don’t want to use it?
Saturn hasn’t been updated in months, is it still an ok option? create-react-app is awesome too, but it’s only for react, so includes no server support.
I would like to combine my client and server into one project, but it’s unclear if it’s a good idea, merging client and server dependencies in the same package. What’s your approach?
I wonder if one can create a package.json file that only scripts that calls executes targets in subproject package.json’s. That would make possible creating a master project for client/server parts.
Why wouldn’t it be an ok option?
Speaking from personal experience, I regularly revive old projects that haven’t had any activity for a few years. No activity may be a sign that the codebase was a deadend, or is buggy, or whatnot. But it can also be a sign that the author set out to implement a certain bit of functionality, completed it, and that the code is stable and ready to use.
For instance… iron:router is still chugging away, rendering pages as good as two years ago. We wanted to extend it, so we forked it, added the stuff we wanted; and now have clinical:router. Easy peasy. No breaking changes; no changing APIs. Get stuff done.
I’m sure Saturn is just the same. In some sort of stable state, and ready for people to tinker with.
Thanks @awatson1978. That may be the case, but I thought I heard somewhere that MDG wasn’t planning to maintain it, and when there’s very little activity, no activity for a few months (no issues, no conversation) then I wonder if the project is dead, superseded by another alternative.
Found this discussion about merging saturn with create-react-app: https://github.com/apollostack/saturn/issues/31
It’s great if people want to tinker with it, but if one wants to use it and doesn’t want to tinker with it, they should know if it’s still in a good condition. The real issue is that JS is in such a state of flux, webpack is now relealsing to v2, lots of parts are moving under us.
Also I was thinking of using CRA or Saturn, but the level of abstraction bugs me. No webpack.config.js. Instead, magic scripts. What’s wrong with “webpack-dev-server”?
There’s over 20 different React starter projects with over 1000 stars. The level of choice in the JS community is kind of ridiculous.
I would give Next.js a very serious look. It uses React.js for the front-end, and has it’s own Router to help with SSR and whatnot. It uses webpack under the covers, has code splitting and SSR out of the box, etc.
The interesting thing is that the next release will allow custom routing and custom server side route handling. This is the last piece missing in my mind to build a really strong front end app ecosystem that works out of the box. You can see it in action on https://conf.crater.io