what can my normie wife do after I die?

[Alan]

I have a fairly substantial accumulation of Ledger data going back to 2002. I rely on it for budgeting (it was invaluable when I was figuring out if I could afford to retire), for reconciling bank and credit card account statements, and -- crucially -- for keeping track of revenue and expenses for my wife's small business, for tax filing purposes. It has been working well for me, and occasionally over the years I have invested some effort in writing helper scripts where needed. 

So I was thinking that everything was all fine. Except recently it dawned on me that I'm not going to live forever. My (young) wife will likely survive me, and then she'll need to do something to cope with finances that I've been taking care of all these years.

Her level of comfort with using computers is at the level of using web apps (e.g., GMail, Google Docs and Sheets, though currently at a quite basic level). I don't think she'll be persuaded to learn how to edit a plain text file and run commands from a Unix shell; to say nothing of learning how to wield Ledger commands and options themselves. Or, at least, she'll not want to continue to do things that way forever.

Maybe I could set up some spreadsheets for tracking of her business finances? Though that might be overly simplistic.

Or maybe she could use something like Quicken?

How can I set things up to ease that transition, especially when it (hopefully) won't happen for still quite a few more years now?

[mb...@mbork.pl]

That's an excellent question. Frankly, I would probably use an LLM tool
to create a GUI front-end to your Ledger setup (including the scripts).
Example: this is an (unfinished, but working as a PoC/MVP) app
I "vibe-coded" for a friend:
https://github.com/mbork/bank-statement-analyzer#.
Note that it even has a Windows installer (and I don't even know how
it works!). Also, it's not really "vibe-coded" in the usual sense of
the word - while I have /no idea/ how the GUI parts work, all the logic
is either written or at least reviewed in detail by me.

It took me about 32 hours over several weeks. Given that I used that
project to learn a bit of Python and spent considerable time actually
/writing/ it /myself/ under Claude's guidance, and that it is probably
overtested (100+ tests for a relatively simple tool), that time could be
driven down considerably if the goal were to write it faster.

Hth,

--
Marcin Borkowski
https://mbork.pl