Sven Slootweg @joepie91@pixie.town

@JimmyB @pezmico "or even a prevalent lack of agency, independence, and spaces for socialisation"

@mynameistillian I think you're spending a lot of spoons (involuntarily) on your shitty living situation right now, and I suspect that that's what's getting in the way of holding a job, more than college

@JimmyB @pezmico

"Why do you think I want kids to have no social contact???"

I said no such thing. I am asking specifically *where and how* you expect kids to socialize. That question remains unanswered here. Okay, "safe places" - *what* safe places? The whole point established in the original post was that there *are* no such places anymore.

@JimmyB @pezmico You "might" get stabbed anywhere at any time. And yet people go outside, they go to places where this is an elevated risk, even. Because yes, social contact is actually necessary, and that has a certain amount of risk.

If there are no social spaces for kids, and you take away their remaining venue for social contact (or never allow them to have it, doesn't matter), then where and how exactly do you expect kids to socialize?

@JimmyB @pezmico "Not giving kids access to them" and "taking them away" are functionally equivalent here. My original point stands.

@nick @researchfairy @derickr Ah yes, Safari, the browser which notoriously only runs on very expensive hardware that is completely inaccessible to a huge amount of people.

Definitely "the most viable option". Mhmm.

@JimmyB @pezmico The problem with this reasoning is that those very phones are often kids' only refuge from the other problems (like a lack of social spaces). So you mostly just end up taking away a coping mechanism instead of solving any problems.

@3TomatoesShort My reason for concluding 'EDS' was basically watching https://youtu.be/3tr1RvErGn8, playing along at home, and getting several full bingo cards in the process (though the suspicions had been there for longer)

@serapath Just a .concat wouldn't be enough because it doesn't create a module scope; that's what the small amount of additional wiring is for, but that's also just strings

@serapath Neither Browserify nor Webpack do code transformation (or at least for Webpack that's true in its initial version). Both just do concatenation.

Both support plugins, which *can* do code transformation, but that happens entirely outside of the bundler.

@serapath Like, to be explicit about this: bundling code does not require *any* kind of code parsing or processing beyond identifying require/import calls, which can be done even if with a regex if you want. Which I believe is exactly what Browserify does, too.

@serapath None of those things (code transforms, tree shaking, minifying) are bundling. Source map generation is only a part of bundling insofar it describes the mapping from many files to one JS file.

Calling code transforms 'bundling' is like saying that Make 'compiles' your code. It doesn't. It just invokes the thing that does, if that's what you tell it to do in the process.

(Notably, switching from bundles to manifests also removes exactly zero of those things-that-are-not-bundling from your build stack)

@serapath Generating manifest files is functionally the same thing as bundling. "Bundling" is nothing more than statically traversing the dependency tree and string-concatenating the results with a minimal amount of boilerplate code to make the references work.

I don't know where people are getting this idea that bundling is some kind of highly complex or objectionable or slow process. It's kind of bizarre.