> On Mon, May 14, 2012 at 02:45:30PM -0600, John Wiegley wrote:
>> Given what we have today, I think the path of last resistance to getting
>> something like this up and running would be to use the Python bridge to write
>> a GUI in either HTML5 or an app framework like WxWindows.  The advantage to
>> HTML5 is that not only could it be accessed remotely, but it would be easy to
>> write mobile interfaces as well.

While I'm all for ending up there, it seems to me like we still have
an opportunity to build a suite of discrete tools:
e.g.:
o ledger-invoice: spits out invoices based on the queried account or
tag or txn(s)
o ledger-bookkeep: Merges a set of new txns into a central ledger
file, makes sure ledger parses it okay, saves it, and checks it into
version control. (This would be the only program that would actually
modify a ledger file)
o various scripts running complex queries on ledger for the
receipt/reimbursables tracking, etc.

I'd love to help develop this, but I don't actually run a business
myself so I don't know whats needed for accounts receivable, etc.. but
I feel like if we first build a set of tools that automate each
different process Bradley has to do, we can abstract them afterwards
for the real gui application that less technically adept people could
use, while still preserving the ability for power users to go in and
work by hand.

Reply via email to