Re: [xml-dev] So maybe ID isn't a problem after all.

[Gavin Thomas Nicol <gtn@rbii.com>]

On Tuesday 13 November 2001 02:10 am, David Brownell wrote:
> Actually, there's the "how to maintain the links" problem too.
> When an author has gone to the effort to define IDs/anchors
> for use in linking, it's a lot more realistic to believe those IDs
> can be maintained over document evolution ... including the
> negotiated transformations that David Carlisle mentioned.

Acutally, the ID-ness here is orthoganal to the issue of link maintenance. 
Unique attribute values are important, not that the attributes be ID's.

> Since that's NOT true of XPath style addressing (navigating
> over a structure that usually changes between incarnations),

Depends on the XPath construct. You can match unique attributes using XPath 
too.