[
Lists Home |
Date Index |
Thread Index
]
Alessandro Triglia <sandro@mclink.it> writes:
> > From: Michael Kay [mailto:michael.h.kay@ntlworld.com]
<snip>namespace problems discussion</snip>
> > I have advocated one change which I believe would alleviate
> > the problems:
> > there should be a lexical representation of expanded names
> > that uses the URI
> > and local name,
>
> That seems to be a "yes" to the last of the questions I asked above:
>
> ------
> When people complain about namespaces, do they mean that
> namespaces should not exist at all? Do they think they are
> useless? Or do they think they should be replaced by
> something else? Or do they have in mind some simple changes
> to the syntax, such as using URI/localname pairs everywhere?
> ------
>
> Joe answered "no" to the same question.
>
> Curiously, most of the responses were related to a question
> that I did *not* ask ("Why do people complain?"). This is
> not what I was asking.
>
> I am not advocating any particular position. I am just
> curious about what kind of solution people have in mind when
> they complain, and whether it would be possible to agree on a
> solution.
One simple thing occurred to me, for much (probably all) of the problem
domain I'm worried about I don't need to distinguish attributes at the
name space level. As such, there is nothing to stop me (for our
internal data for our apps.) from using an attribute to distinguish
"namespace". That fixes my architectural/structural design problem: I
no longer have to commit to the name space at design time I can just add
it in later if and when I run into a name conflict. In our case it
would be easy enough to just say attribute "ns" is reserved, and convert
it into a real name space whenever things went external...
|
