On graphs and equivalences

[Shawn Steele]

> To be fair, we don't need to go that far.  The goal of this WG was to
> "Internationalize LDH".  Since LDH is case-preserving and
> case-sensitive, one could (I do not, but one could) argue that some
> kind of case-preserving case sensitivity is desirable for an
> internationalized version too.

That's a very analytical view.  "Case-preserving" provides 99% of the linguistic functionality an English speaker could request.  I'm not sure what else would be interesting for English, though often we ignore diacritics.  So one could instead argue that "99% linguistic functionality" would be an appropriate bar :)

We have to remember that the people that actually use the DNS system aren't network engineers.  They're marketing execs and other "normal" people.  A perfect computer solution isn't necessarily the right solution for the target audience.  (or at least my target audience, which isn't restricted to computer geeks).

-Shawn