Personally, I live the Office license my workplace pays for untouched, and use LibreOffice on my work Windows computer instead.
Yes, it's slow and bloated. But it's comparably faster and leaner, and it doesn't use undocumented APIs to take resources away from everything else running on the same computer and make every other thing unusable.
And yeah, calc lacks features when compared to excel. So, avoid spreadsheets for complex problems.
I haven't used the office tools on Linux for years, but since they are based on the same 30 year old idea of an office suite, would you expect them to be much better? I don't think it is productive to choose among equally antiquated and poor quality alternatives. It isn't going to advance the status quo.
In terms of word processing (which is perhaps an archaic term by now) I would ask people to look at what Visual Studio Code is. A rather minimal, skeletal, code editing platform that derives nearly all its value from the extensions people make for it. There are lots and lots of editors and IDEs. But extremely few of them serve as platforms. As the infrastructural basis for creating applications.
Yes, there are IDEs that are possibly marginally better at editing, say, Java or Go code. But VSC is pretty good at almost every language that is in common use today. And it manages to compete pretty well with more specialized solutions. It does this because an editor that does 90% in all the languages you use is far more valuable than switching between two editors that perhaps achieve 95%.
Word, and its open source counterparts, are antiquated and obsolete. I don't think the field can be advanced by building word processors that are just iterations of 30 year old ideas. Yes, you can probably extend them, but people don't. You have to understand what it is that makes some pieces of software work as platforms (like VSC), and why other pieces of software do not inspire people to build on them.
I think Microsoft should reinvent Word as a platform that is designed to be extended and that is easy to extend. I would then release the base software platform as open source. Much of the functionality that resides in Word today I would move to paid extensions - including useful bundles of extensions. This way Microsoft would retain its revenue stream, and I wouldn't have to deal with all of the crud Word contains.
I would also create a marketplace for both paid and free (open source) extensions. Which in turn would make the product more valuable (even though the base product is free). Because other companies and people invest in it and have a shared interest in its health beyond mere existence.
Of course, not only Microsoft can do this. Anyone could create an editing platform. But it would have to be someone with a bit of money who can spend perhaps 5-6 years supporting the effort to see if it takes off. Maybe it does, maybe it doesn't.
One reason I see this as perhaps the only way forward for this class of application is that I'm doing some work for a company that manufactures physical products. A would-be advanced user of office automation tools. This kind of business has a very complex document structure where there's a vast hierarchy of thousands of documents that goes into every project and even spans projects. Doing this with Word, Sharepoint and whatnot is complicated, fragile and requires a lot of work. It doesn't work very well. It also means you have to memorize a lot of procedures. This could have benefitted from very narrow, domain specific tooling. Including LLMs that allow you to ask questions with context derived from sources other than the Word documents. Yes, Microsoft is trying to stuff this into their products, but it isn't actually all that useful because it is generic. It is never going to support what our customer needs.
I don't think Office, LibreOffice etc are the right kind of tools. They are children of the 1990s. We have better starting points today and better technology. It is time to re-think this.
I love this idea, and agree with your assessment of office tools in general. You almost make me want to try it, but it's evident the size this body of work would take.
