On Mon, 24 Apr 2000 02:30:15 CDT, the world broke into rejoicing as
[EMAIL PROTECTED] said:
> It's been rumoured that Christopher Browne said:
> >
> > Given freedom to choose a way of representing this, I'd think the
> > Right Way to be to have, for the stock, an indication of splits that
> > have taken place, in much the same way that prices are tracked.
>
> Imagine a 2-for-1 split ...
>
> One could create a balancing transaction by 'selling' 100 shares at
> $100 and 'buying' 200 shares at $50.
Unfortunately, that establishes a taxable event, which _hasn't_ taken
place.
What has occurred is just the same as a price change, and price changes
don't result in there being new transactions.
> By putting both ends of this transaction in the same account, we'd be
> done ... except for a few bugs ...
>
> -- first, the current register code would display this transaction
> twice, (which is a bug), leading to no end of confusion.
>
> -- next, we have that sign-reversal problem debated in earlier emails.
> There'd have to be some care to make sure the sign didn't get
> accidentally reversed: there might need to be special case code
> for detecting when both ends of a transaction are in the same
> account.
>
> -- finally, this might introduce problems with the cost-basis code.
>
> On the other hand, this seems to have less of an overhead than
> introducing 'yet another field' into the engine, and tracking the value
> of that field ...
The model of setting up a sale-and-repurchase transaction to represent
a split _is incorrect_, as no such transaction takes place, any more than
a sale-and-repurchase takes place when the price of a stock rises from
$50 to $75.
The alternatives are somewhat ugly; it seems to me that it would be a bad
thing to oversimplify this. Things should be as simple as _necessary_,
not as simple as _possible._
--
"Other than that, Mrs. Lincoln, how was the play?"
[EMAIL PROTECTED] - <http://www.hex.net/~cbbrowne/lsf.html>
--
Gnucash Developer's List
To unsubscribe send empty email to: [EMAIL PROTECTED]
