We are working on backend for life insurance. They have a separate pricing team who creates calculations for different insurance products. The results are usually presented as excel files with heavy scripting inside. Really heavy! Have you seen 100mb excel files? I did!

On the funny note: as powerful excel is, it cannot open two files with the same name from different folders! Or at least my version can't.

I used to work on Excel extensibility. I’ll never forget when we were talking with one of the top 5 insurance companies in the US and they showed us this huge VBA macro and told us they processed the top 10% of their claims using this file. That moment made me realize Excel powers at least 10% of the world economy.

Oh yeah, because references to the book name would conflict. If you open a new instance of Excel (e.g. shift-click the taskbar icon), you can open one file in each, but they can't reference each other.

I have seen 600mb+ geodata based, script bloated, slow as hell, make your MB Air commit sucide type of excel every day for 4 months, working on a project.

Excel is also used and abused. In finance and insurance, often Excel isn't used as a spreadsheet and visualization application. It's used as a database and application engine.

This is really bad for a lot of reasons. Of course it's painfully slow, but it's also incredibly brittle and foot-gunny. Excel IS NOT a competent database engine or application engine. It makes JS and C++ look sane and safe.

Excel shouldn't be switched out to LibreOffice. It should be switched out to a proper application with a proper database. What, finance bros don't know how to navigate a database. Tough fucking luck! In the 70s, secretaries could do that. They better figure it out. Because these existing "systems" are a disaster waiting to happen.

I mean, it's not as if Libreoffice was created yesterday. Its predecessor OpenOffice is now 25 years old. Your question becomes: Why did you choose the expensive version where you never know what it will cost next year in the first place?

Concerning

> Why did you choose the expensive version

Big companies have (sometimes hard negotiated) volume contracts with Microsoft, which makes Excel much cheaper to them than to, say, small companies. Thus Excel is not really expensive for them.

Concerning

> where you never know what it will cost next year in the first place

For open source software there exists a similar risk that you don't know into which direction the product will develop.

In the past, Microsoft has been quite reliable in keeping backwards compatible, and continue selling office for decades.

In my observation, the zigzag course that Microsoft starting doing with Windows (but is now also doing with office), and, relatedly, deviating from the course of being very insanely dependable in delivering the software that companies need from them, is what by now got big companies at least have a look at what possible alternatives to Microsoft products could be.

> For open source software there exists a similar risk that you don't know into which direction the product will develop.

You know what's really interesting to me about this argument point?

It is actually the proprietary solution that is at risk of this, and we feel it daily. The next version of Microsofts own flagship product (Windows) is nearly universally denigrated, but people are forced to upgrade.

With FOSS, there's significantly less risk, if the product changes direction you and your other company friends can just use old versions or in the worst case.. fork it.

As a counterpoint, Microsoft is in the process of discontinuing Publisher and plans to remove it from M365 installations next year.