Re: Len Silverston's Universal Data Models sanity

From: Pascal Damian <pascaldamian_at_icqmail.com>
Date: 7 May 2004 22:34:06 -0700
Message-ID: <6bd4a4d3.0405072134.66f63029_at_posting.google.com>


Leandro GuimarĂ£es Faria Corsetti Dutra <leandro_at_dutra.fastmail.fm> wrote in message news:<pan.2004.05.07.12.14.36.209211_at_dutra.fastmail.fm>...
> Em Fri, 07 May 2004 01:56:28 -0700, Pascal Damian escreveu:
>
> > In what way does the Brazilian accounting model differ from
> > US (and other parts of the world)?
>
> > Surely the basic concepts and entities
> > of double entry, debit/credit, accounts /chart of accounts, transactions,
> > etc are the same?
>
> I don't know what's a chart of accounts, the rest of the
> concepts apply.
>
> > What does the Brazilian accounting model look like?
>
> I never worked with US accounting, I know it is sufficiently
> different that US companies in Brazil keep two sets of accounts, one
> to show to headquarters, other for the legal obligations in the
> country.

Yeah, all international companies will need to maintain multiple books. But these are due to the different accounting _rules_ and _practices_. A sufficiently generic accounting model will be able to handle these fine, as the business rules itself are not embedded inside the model. I had only read Silverton's 1st edition of the Data Model Resource Book, and I remember his accounting model is pretty generic (transactions, transaction detail, chart of accounts, accounts, account types).

Things like:

- how, when, how much an asset should be depreciated;
- how to calculate income tax, sales tax, etc;
- when should an income be earned, or posted, or recorded;

should belong in the business rules, _not_ in the model. Otherwise the model will be overly complex, very brittle, and needs to be modified every now and then, for every company.

--
Pascal
Received on Sat May 08 2004 - 07:34:06 CEST

Original text of this message