> > I guess I've always meant to do the automation but have never gotten to it > as I haven't seen enough actual success stories with it for basically the > reasons you just posted (I don't typically read the mailing list). I've > been through too many "works for me" recomendations (for other > technologies) that don't work to trust it at face value, so I would have to > spend significant effort to test and validate all these different > experiences. I am also in the US with some credit unions etc. and I > couldn't even get them to connect with Mint even though I had friends that > could do the same. Lesson learned there was just to take ownership of it > and no one will do it for you.
All excellent points and they mostly line up with my experiences too. BTW, one exception: getting your ingest to directly download from your bank is is actually only a small part of the picture. This is why I recommend having that part be one of the last things you automate. See here <https://reds-rants.netlify.app/personal-finance/what-should-you-tackle-first/> . > These days I just don't generate enough "1 USD coffee" transactions to > make me motivated to automate them, part of which I credit to painfully > hand entering all these transactions :) It also forces me to look at each > transaction to know if any are fraudulent. For a beginner to beancount or > just someone thats financially unsophisticated I think its valuable to do > it by hand. > Agree again: doing it by hand first is very educational. > But all that said, yes I would like to automate the tedious parts. I'll > try to do a little bit more investigation, but I would like to see a "Do it > this way" kind of guide that isn't just "works for me" with a reasoned > explanation of all the problems, gotchas, etc. in detail so I know what I'm > getting into and where something may or may not work. > You captured exactly why I wrote the Five-Minute Ledger Update <https://reds-rants.netlify.app/personal-finance/the-five-minute-ledger-update/>. Here is the mailing list thread <https://groups.google.com/g/beancount/c/_NclCTXaExs/m/rBdKP4U4AgAJ> if you haven't seen it. For this particular problem, sharing code = "works for me". The guide above is the "do it this way". I'm happy to hear about what helped you and what didn't from the guide above if you read/use it. On Friday, December 31, 2021 at 6:30:49 AM UTC-5 redst...@gmail.com wrote: > >> >> I agree with Blais that there really isn't a better solution to all of >>> this. I think there is a misconception that because it is conceivable that >>> all of your transactions can be automated (and they are for individual >>> banks & institutions) that it should be easy to automate or that there is >>> already a solution. It sucks to have to do it all manually, but the pain of >>> entering hundreds of "Expenses:Coffee 3.00 USD" transactions now makes me >>> think twice about it! (or to just use cash and book them all in a single >>> pad entry :P) It also has really made me appreciate the complexity of some >>> financial instruments, which in my case has made me less afraid of them >>> since I know that I can keep track of them all as I want to start doing >>> crazy things to make money :) >> >> >> Indeed, close to all of your transactions can be automated close to all >> of the time. ofx rarely breaks and is indeed highly automatable. csv may >> take longer to develop and might occasionally break. Getting it from 95% to >> 100% is usually not worth it in my experience. But getting it to to the 95% >> point is extremely valuable, and the point of the 5-minute ledger update. >> >> PS: I haven't entered "hundreds of 'Expenses:Coffee 3.00 USD' >> transactions" or similar in at least half a decade. Nobody needs to do this. >> >> >> I really can't imagine having to do all of the importers myself and >>> maintain them since the banks are always changing and no two are alike. I >>> think its just easier to do by hand as of now since I know after working a >>> few hours I will have a product, rather than spending a few hours >>> automating and it just sucks at the end anyways. I would definitely pay >>> someone to provide up to date importers and a UI for quickly categorizing >>> them (since the banks never get this right for my purposes). >>> >> >> With the ingest frameworks available (including mine), the average user >> should be able to recoup their cost in say 2-5 import cycles. If you're >> like me and like to import several times a month, there's a tremendous ROI >> in automating. And no, it's extremely far from "just sucking at the end >> anyways". >> >> A whole bunch of people have shared importers. A pypi-like system to >> list, install, and manage importers would be awesome. Without that, most >> users need to put in some time up front to find the importers they need. >> >> Of course, all this depends on where you live and whom you bank with. My >> personal experience is as a user in the US. >> >> I'm curious as to where your impressions above come from, given mine is >> far from it. Perhaps if you could share whom you bank with and what >> specific problems you've had writing importers, that would help. >> > -- You received this message because you are subscribed to the Google Groups "Beancount" group. To unsubscribe from this group and stop receiving emails from it, send an email to beancount+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beancount/36dc721c-47e3-4cc1-a1e9-7ba8ecd98174n%40googlegroups.com.
