Re: Parents disinherit their children

Subject: Re: Parents disinherit their children
From: crism@xxxxxxxxxxxxx (Christopher R. Maden)
Date: Wed, 13 Oct 1999 15:28:40 -0700
[Elliotte Rusty Harold]
>However, I'm concerned about the logical inconsistency in this
>statement as currently written. In common usage, both technical and
>genealogical, the statement that A is the parent of B clearly implies
>that B is the child of A. Why is this common understanding of
>language broken here? Is there anything that can be done to fix it?

It is a little confusing, but it seemed like the best compromise.  It
seemed clear that attributes and namespace nodes were distinct from
children, but it also seemed clear that it would be confusing and
cluttering to have two distinct axes to return to the origin of a node.
DSSSL uses the origin property for this, but that name didn't fly in a
language intended for non-mathematicians.  The XLink WG really didn't like
calling it "parent", and (I believe) wanted two distinct axes, but the XSL
WG felt that the confusion of parent and child being non-symmetric was less
than that of having two axes.  This is especially true when considering the
shortcut syntax: it's not so bad to have

child::para/parent::*
attribute::class/origin::*

but unless you break the formalism, you can't really have the same
shortcut, so you'd end up with

para/..
@class/@..

or

para/..
@class/origin::*

or something like that.  I think most people won't ever use the full axis
names, and will just understand that .. means up one (of anything).

-Chris

--
Christopher R. Maden, Solutions Architect
Exemplary Technologies
One Embarcadero Center, Ste. 2405
San Francisco, CA 94111



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Current Thread