@ix5 I don't understand why this is needed. Aren't we using Webpack for this reason? Can't we write our code in modern syntax and let webpack convert it for us? This is how every app and libraries does these days, I don't get why we can't do that. Disclaimer: I haven't dug into the codebase, it's just a random idea.

Re: Your questions:

Aren't we using Webpack for this reason?

The move to webpack was mostly to enable testing via Jest and associated tools. It's a massive dependency carrying megabytes of transitive dependencies and I'm not happy about it, but it was a necessity of moving forward.

Can't we write our code in modern syntax and let webpack convert it for us?

Adding the following to package.json:

diff --git a/package.json b/package.json
index e481c7f..4049ac8 100644
--- a/package.json
+++ b/package.json
@@ -13,6 +13,9 @@
   },
   "devDependencies": {
     "jest": "^27.5.0",
+    "@babel/core": "^7.17.8",
+    "@babel/preset-env": "^7.16.11",
+    "babel-loader": "^8.2.4",
     "webpack": "^5.68.0",
     "webpack-cli": "^4.9.2"
   }

balloons the size of installed packages from:

$ npm ls -all | wc -l
830

to

$ npm ls -all | wc -l
1159

package-lock.json swells from 8363 to 10901 lines.

$ find node_modules/ -name '*' -print0 | wc --files0-from=- -l
[...]
1057313 total

¹

While the current state is already very unnerving to me, adding even more cruft pushes the dependencies to a level that is not reviewable by humans any more.

This is how every app and libraries does these days, I don't get why we can't do that.

Yes, it seems that's the way the wind is blowing these days. But just as a reminder:
For 0.12.6, released just a few weeks ago, we had this:

    "almond": "^0.3.3",
    "jade": "^1.5.0",
    "requirejs": "^2.3.6",
    "requirejs-text": "^2.0.15"

Taking jade out of the equation (because we dropped the templating), we are left with just three packages which each have no further dependencies.

I'd like to underline that this PR only restores the status quo of syntax that was already present just a few weeks ago.

If this project is to move towards a new standard of JS syntax (e.g. ES6 and getting rid of the drop-in promise "polyfill"), it should be because someone can demonstrate the actual need for this. There has been a lot of enthusiasm for new and shiny DevOps tools, helper bots and the like, but I have yet to see anyone actually contribute any new Javascript test or an improvement to the client in recent weeks.

If the lack of "modern" syntax is actually preventing people from contributing, then I'm all for lifting this "restriction". But until there's good evidence, I see little reason to throw so many internet users under the bus (or to introduce a total of more than a million lines of code in dependencies).

An updated data point:

Since making the Jest dependency an "optional" one ¹ in d797bed, the amount of total installed packages is 166:

$ npm ls -all | wc -l
166

With babel dependencies added, we get 589:

$ npm ls -all | wc -l
589