Here's an example: Imagine if your bank account had a profile, almost like a social network, and you could go to someone else's profile and quickly see recent transactions you've had with them. I'm not thinking of something like a credit union, rather something quite different, and I really feel there's lots more that can be done with banking from a tech pov
How revolutionary is that, exactly? Any decent online banking service should allow you to see your payment history and filter it by the entity you paid, even if you have to export to csv to run the filters.
Of course many people would value a a pretty UI layer on top of that with categories, notes and charts and graphs that offer much more interesting graphical indications of which people and companies take up the highest portion of your monthly spend, where your bills are going up and which shops have you not visited in a while ... but then you find yourself in the position where you're really trying to build an open source alternative to mint.com rather than a bank.
...Which actually seems like a much easier ambition to realise, not to mention a niche which the likes of GnuCash have started to fill.
I think it'd be the experience of that rather than the actual feature. Like how many non technical people are going to straight out of the box know what export to csv is then be able to run filters on it? People are much more familiar with my proposed way. It's note just the UI, it's the way it's done. Paraphrasing Steve Jobs, the design isn't the UI, it's the way it works.
Mint is great, but it's very much read only. A read and write setup, if you get the anaology, would surely be much more powerful.
