ID for language-invariant strings

[David Starner]

On Sat, Mar 15, 2008 at 10:06 AM, Peter Constable
<petercon at microsoft.com> wrote:
> I think any information-technology engineer here would acknowledge
> problems of inappropriate overloading. We've got two camps saying
> my concept should be handled by overloading an existing code element,
> but they are arguing for different code elements: zxx versus und. Doesn't
> that suggest to anybody else here that just maybe those people don't
> agree because this is an instance of inappropriate overloading?

No more than two people disagreeing about whether to tag "Gift" en or
de. If you want to express the concept that here's text in some
language, we don't really know or care which, und is your tag. If you
want to express that you have an arbitrary chunk of text that's
serving as a tag and not real text in a language, then you tag it as
zxx.

The closest think I can find on quick search as to a specification you
gave on what this new tag should mean is
> That's leaving me a bit inclined to think that we should establish a
> special-purpose tag for the language-invariant semantic.

One of the problems I have with that is that either it's a name, in
which case I don't see this as being any more language-invariant than
any other name (and if you look at Everson's page, or the interwiki
links on <http://en.wikipedia.org/Coca-Cola> you'll see that the names
frequently do get changed on changing languages), or it's an arbitrary
stable tag that happens to be displayable as text, in which case zxx
works just fine.