gnucash budget

Chris Browne cbbrowne@localhost.brownes.org
Fri, 27 Jul 2001 15:33:42 -0500


David Reno wrote/forwarded:
> >  -----Original Message-----
> > From: 	David Reno  
> > Sent:	Friday, July 27, 2001 11:01 AM
> > To:	'gnucash-devel@lists.gnumatic.com'
> > Cc:	'jsled@asynchronous.org'; 'bryanlarsen@yahoo.com'
> > Subject:	Re: gnucash budget
> > 
> > 
> > Regarding "types" of budget items:
> > I think finding a general form for budget items is best.  I think any 
> > distinction between things like number of occurrences should be treated as
> > an 
> > attribute, not as a profoundly different type of item.  To support this,
> > lets 
> > consider the "types" that have been discussed so far:
> > IRA contribution:  $2000 budgeted to be spent over 1 calendar year period.
> > To 
> > be budgeted yearly, recurring.
> > 	budget_item: (comparison expense category =IRA Contribution),
> > 	(item start date=1/1/2001),
> > 	(item end date=12/31/2001),
> > 	(amount=2000)
> > Holiday Gifts:  $200 to be budgeted to be spent over 2 months.  To be
> > budgeted 
> > yearly, recurring.
> > 	budget_item: (comparison expense category =Gifts Given),
> > 	(item start date=11/1/2001),
> > 	(item end date=12/31/2001),
> > 	(amount=200)
> > Savings Goals:  Increase savings by $2000 this calendar year.  To be
> > budgeted 
> > yearly, recurring.
> > 	budget_item: (comparison expense category =?),
> > 	(item start date=1/1/2001),
> > 	(item end date=12/31/2001),
> > 	(amount=2000)
> > Toy:	$500 to be spent next week.  Not recurring.
> > 	budget_item: (comparison expense category =Entertainment &
> > Recreation),
> > 	(item start date=7/29/2001),
> > 	(item end date=8/4/2001),
> > 	(amount=8)
> > So my point about different types is that if we view them as being more
> > similar 
> > than different and treat the differences as attributes, we may have a
> > simpler 
> > model to work with.

There is some basic "class" information, with variants that specialize on it; 
this stuff looks a _WHOLE_ lot like the sort of thing where the "generic 
functions" approach of GOOPS would represent a useful "attack" on the problem.
--
(concatenate 'string "cbbrowne" "@acm.org")
http://www.ntlug.org/~cbbrowne/
He doesn't have much of a reputation, or so I've heard.