The benefits are so obvious that they preclude any supporting evidence or research. They are more or less axiomatic/dogma.
Wild take - maybe the pace of introducing new changes has not increased _because_ the "benefits" are not so obvious after all.
But in every framework I've used, the suggested way isn't how a technology is used in reality, in production. The tutorials are almost like a different framework entirely. Years ago as an Android dev the difference was shocking between what tutorials taught you and what was done in practice.
This doesn't technically detract from the author's point but it makes it moot - you just build in the current best practice way or the way that suits your needs, and that's how it often is with languages and frameworks.
Maybe that mismatch between how people are taught initially, or how the framework is intended to be used, is an inefficiency, in which case those who design frameworks should take note.
In the case of Rails I think "the rails way" is appropriate for certain style of apps, and not so much for Shopify etc scale apps.
In any case, I would not advise anyone to take any advice in web-development from someone who does this.
P.S. My browser is quite fine, release age wise, not that it should matter.
And this is accelerated by design patterns that blow up the amount of components used. Then there's the problematic gems such as Trailblazer that cause more problems than they solve, while being difficult for both humans and AIs alike.
Reminds me of the quote from Fred Brooks: "Show me your flowchart and conceal your tables, and I shall continue to be mystified. Show me your tables, and I won't usually need your flowchart; it'll be obvious".
Service Objects et al are flowcharts. Rails-way is tables.
https://shopify.engineering/shopify-monolith http://sporto.github.com/blog/2012/11/15/a-pattern-for-servi...