It's crazy how we've gone from a seemingly inescapable monopoly on office software in the 2000s to it being optional. I remember working at a previous job where everyone used Google Sheets by choice and they would raise their eyebrows if you spun up Excel.

Lots of companies don't have internal IT departments, and regularly get swindled by 3rd party IT companies (read: licensed Microsoft/Google salespeople) to switch over their entire systems from one or the other, productivity be damned.

Yet do they use ancient copies of the software that broadly still performs the tasks people need of them? No.

Yes, actually—I have a VM reserved mostly for 16-bit software.

Do they theme their system to look like the oh-so-superior Win98? No.

Yes, actually—the Windows machine I'm forced to use for work restores as much of that aesthetic as practical, sometimes with the help of third-party software. My main home machine features a Linux DE whose appearance is largely the same as it was circa 2005 and whose development team is dedicated to keeping that look and feel.

Some of us do put our money where our mouths are, although I admit that isn't universal.

It's true that some level of padding is necessary in a UI, but the amount present in contemporary design is way too large for a system using a traditional mouse or laptop touchpad, which are capable of small, precise movements. Touchscreen-friendly design is best saved for touchscreens, but people don't want to do the work involved to create multiple styles of UI for different hardware. I've never encountered anything touted as "one size fits all", whether it be a UI or a piece of clothing, that actually does fit everyone. At best, it's "one size fits most", and I'm usually outside the range of "most" the designers had in mind. At worst, it's "lowest common denominator", and that seems to be the best description for contemporary UI design.

Yet do they use ancient copies of the software that broadly still performs the tasks people need of them? No.

This just means that functionality and interoperability criteria are more important than usability. They are - you can't just exchange docs with a person using a modern office suite, while you are using WordPerfect 8 for Linux.

This is the opposite of confirming your argument about UI\UX, because this means that UI\UX are order of magnitude less important in making the decision.

And it's obvious, I swear, some people haven't been taught that arguments are not intended to support their group or hierarchy, you can't do that with cheating in arguments anyway. They are intended to find out truth, make both participants richer than before.

Do they theme their system to look like the oh-so-superior Win98? No.

That's simply because they "theme their system" to look as they wish and they don't have to stop with Win98 or Win2K.

But in a "one size to fit all" situation those are still obviously superior.

Ergonomics is not a matter of opinions, there's plenty of research since the fscking world war two. Different controls should have different colors, shapes and textures. It's a scientifically proven statement. Proven with human error stats and time to do a task stats.

Padding controls and indicators with space can be a good thing, but no modern designer is doing it right as far as I'm concerned. Because it's not about making panels half the screen, it's about different groups of controls being clearly separated by that space and padded for focus, and space being used proportionally to importance.

They've all heard something of it, but haven't learned the actual thing.

Older UIs were usually (often, but not always) made with respect to ergonomics.

thank god modern UIs aren’t inconsistent, cramped and cluttered like this

Our ideas of all three things seem to be diametrically opposite. For me older UIs seem ordered, compact and correctly accented. In general, it's not always true - say, I like the appearance of old KDE (2-3), but not sure if I'd use it daily, for example (neither I would modern KDE).