[sc34wg3] Problem with wildcards

[Robert Barta]

On Thu, Mar 05, 2009 at 04:26:54PM +0100, Lars Heuer wrote:
> Hi Robert,
> 
> [...]
> >> This makes sense to me from the point of view that if the author hasn't
> >> created an item identifier explicitly through assigning an ID, it seems
> >> wrong to imply one.
> 
> > I'm worried that we now create a two-class system between item
> > identifiers, ones which the author has explicitly wished for and those
> > which spring into life.
> 
> Since the introduction of wildcards we have this. ?, ?foo and [-"foo"]
> generate automatically a predictable item identifier otherwise we
> cannot ensure that the topic has an identity.

Well, we have/had _different_ ways of creating them. And that is/was good.

What I am worried is that now we have _different_ ways of using
them. I (as implementor) have to memorize this somewhere and I (as CTM
author) have to be aware of that.

Context => Not sooooo good. (Read: very, very bad)

\rho