You're thinking like an SV webdev who's largest concern is to onboard as many new users as possible so you can flip your startup.

Office is a professional productivity tool that's been around for decades and is used by millions of people for their entire workday, every day.

Learning to use a new tool takes a tiny fraction of the time you're going to be using that tool. One should make the tool best at actually doing what it's going to be used to do.

I'm neither in SV nor a webdev. I just remember how many ordinary users hated the ribbon when it first came out. Presumably they adapted over time; I can't really speak to whether Microsoft would have been better evolving the existing menu system. Personally, I could take it or leave it--and at least intellectually the arguments for the ribbon made sense to me.

Ordinary users hate change. They always do, regardless of how successful (or not) the changes are later judged to be.

This doesn't argue against change so much as argue that the costs need to be worth the disruption - you don't move houses over a carpet stain, and you don't rebuild your UI lightly, either.

Generally, though, you shouldn't optimize GUI:s for new users but for intermediate users.

[0] Cooper, "About Face: The Essentials of User Interface Design"

That should be a tenant of UI design, especially for any kind of productivity software.

Think of the Blender, GIMP or Sony Vegas GUIs. Daunting the first time you open them, but after you've successfully finished a few creations you'll begin to appreciate their power.

For any serious work, overly-simplified GUIs are quickly exhausted. They may get a pass for IM or social media apps, but I can't think of many other useful applications.