eliuuk
commented
8 years ago Bundling should be a close to the top as well. For those that don't know what bundling is, bundling provides a way to package all the css and all the js into one tidy file. That translates into faster load times for the end user, but that also means that developers don't have to worry loading order for each script, thats all handled by the bundler. As far as the current code base is concerned, that would mean ripping out the current module loader and replacing it with something like Browserify. At a quick glance that looks like it would probably be a fair amount of work. I think, though, that the gains would would be well worth the pain of a conversion.
I'm not someone who works with js or css, so apologies if my opinion doesn't work! Bundling is something that I'm not completely hooked on with. This is because the .js files in the scripts folder have thousands of lines. For a content creator, scrolling through thousands of lines just to add an airport to be loaded would be slightly annoying. But, if this is something that's crucial, then it's up to you!
n8rzz
erikquinn
Is there any need/desire from the developers here for more tooling?
Things that are lacking in the repo that I think might be beneficial:
Immediately implementable
Long term improvements
Is there any reason a custom module loader is being used? Was that part architected before module loading was even a thing?
I think at the very least there should be some code quality tooling like jshint/eslint and especially an editorconfig file.
Bundling should be a close to the top as well. For those that don't know what bundling is, bundling provides a way to package all the css and all the js into one tidy file. That translates into faster load times for the end user, but that also means that developers don't have to worry loading order for each script, thats all handled by the bundler. As far as the current code base is concerned, that would mean ripping out the current module loader and replacing it with something like Browserify. At a quick glance that looks like it would probably be a fair amount of work. I think, though, that the gains would would be well worth the pain of a conversion.
I think making a transition to ES2015, or es6/es7 as it's sometimes referred to, should definitely be on the radar (pun very much intended) too. ES2015 brings a lot of fun new toys to javascript and makes it much more fun to write javascript. Not mention making the code much easier to reason about. The downside is that it does require some tooling, though it's not that hard to implement and if something like browserify were implemented first, making the switch to es2015 would be much easier.
Most of the changes above involve big changes in folder and file structure that could have big impacts on other developers and the work that they may be doing. But, again, I think there are some substantial gains to be had here.
I'd be curious to see what interest, if any, exists for any of the above. Or maybe there is something else that I didn't list above that some of you would like to see implemented?