> Can or should we allow for different levels of "mungedomain" behavior?
> Right now I find mungedomain mildly useful, but with unrealized potential.
> I'd like something that says, in effect, strip all addresses to 'n' (2)
> domain components, then compare the addresses.
This is a nice idea, and I agree with the intent; however, we need to be
careful of the metric we use for accepting a addresses as having a
"close enough" match.
Consider the case of the domains "silvaco.co.il" and "datasrv.co.il".
".co.il" is the domain for commercial (".co") sites in Israel (".il"),
"silvaco" and "datasrv" are domains of particualar countries. Clearly
we don't want user at one company to subscribe or (perhaps more
importantly) unsubscribe another just because they happen to be
in the same country.
References:
|
|
