I feel like GNUCash splits the difference. You still get a GUI and reporting tools, but you also get the benefit of local reasoning that double entry accounting provides: as long as every transaction is balanced the entire book is balanced. And the data model supports transactions between more than two accounts, which is extremely useful.

My preferred backend is SQL based, so it's a relatively amenable to inspection and custom tooling.

In fact, let’s just use Excel for everything.

Throw away all other software, you don’t need any of that.

Excel is the command line of the data manipulation world.

Except when Excel mangles your genomics. Or dates.

Excel is the worst thing ever happened to data scientists.

If sciencist knew awk/perl + gnuplot and basic C, most of the errors would't happened.

From HN: Scientists rename human genes to stop MS Excel from misreading them as dates

https://news.ycombinator.com/item?id=24070385

R is pretty good to boot, and has been around since the 1990s too. I would’ve said that Excel is the assembly of data manipulation and dplyr is the command line. The higher level of abstraction and the guardrails help.

For my own personal stuff I do just use spreadsheets. Copy paste and some super hacky SQL queries, done. Double entry bookkeeping is total overkill for most people's home budgeting.

I do use ledger if it's for other people's money though, cus the liability of getting things wrong is higher.

This, but with SQL instead of Excel.

Or both! Stuff the SQL into the spreadsheet, shit works great for personal budgeting stuff