The interactive coding with global persistence reminds me of Smalltalk. It’s still around (see: Squeak and Pharo) but hasn’t gained popularity. I suspect a big reason is that it’s hard to make complex software without it becoming unorganized and brittle (though another factor may be lack of industry support partly caused by bad luck and timing).

Also, the author’s company and software (http://frontier.userland.com) seems to be focused on web development. Ironically, web development today has live reload and a scripting language (JavaScript), which has led to unorganized and brittle software partly because of its “scripting” (e.g. weak types, hacky OOP).

Despite the aforementioned flaws with scripting, I actually agree with the author that it should be promoted much more. I’d love to see something like Pharo or a modern version of UserLand software (with more features) become mainstream, because it would empower non-programmers and hopefully lead to more creative software (see: Adobe Flash).

However, I disagree with the article’s macOS and iOS framing. First-party lock-in means even if Apple develops a great scripting framework, unless they make it run on other platforms, it will be less popular than inferior solutions like React Native and Flutter (which already have live reload and scripting and target iOS). Instead, I think it should be a third-party framework, like React Native and Flutter, just with more interactivity and better macOS and iOS support (mainly increase reliability and feature parity with other platforms; the framework would probably have to simpler and targeted for smaller apps to do well both interactivity and cross-platform support).