I think GNUCash is in the Unix mode of "do one thing and do it very well". For me, that "thing" is enforcing double-entry accounting, although there are a coupl

I think GNUCash is in the Unix mode of "do one thing and do it very well". For me, that "thing" is enforcing double-entry accounting, although there are a couple other things that distinguish it from a spreadsheet for me. I can see how you might want or expect a piece of software that handles everything, but in the GNU ecosystem, the traditional user-land pattern of use is to have many pieces of software, each of which does one thing, and to use them in turn as needed. This might place more cognitive burden on the user (vs. Windows), and also requires pieces of software you can start, use, and close quickly (again, vs. Windows, and GNUCash is a bit wanting in that regard...) but it also allows you more freedom and flexibility in doing your work. So, instead of embiggening GNUCash, I would say that your desired actions would better be served by separate pieces of software: - Bank hook-up: Something from your bank? This one is tough, as there is no widely-used open standard for manipulating your bank account over the internet, AFAIK, so you'd need bespoke software for your bank. - OCR data ingest: Something like Ocrad with post-processing into a format GNUCash can use. Some lex/yacc wizard could probably do the latter part easily. - Automatic tax calculation and burden reduction: A FOSS version of TurboTax? This is a good example of why the Unix philosophy is applicable, because my taxes in Canada are simple enough that I can handle them myself. I don't want a big tax-optimisation program tacked onto my bookkeeping software which I won't use. - Expense comparison: I don't think closed-source accounting software does this, but GNUCash does allow you to make budgets. Collecting the budget for a "typical person" (of your country and lifestyle) is a little out of scope, and might be better served by a market research agency or something. Sadly, there will probably never be a GNU World Census. - Mobile app and cloud storage: This one is a little tough. You want it to sync up between your mobile device and (presumably) desktop. You could host a Seafile or OwnCloud instance on your desktop, but I'm not sure if that would let you sync them up anywhere. iCloud-style storage would require server infrastructure, and that usually requires an organization running off profit or donations to host and maintain the infrastructure. IMHO, we don't really have a good answer for the issue of "who pays for the servers?" in FOSS. Anyway, we are getting beyond a software/GNUCash issue and into a hardware/infrastructure issue.