Relationships between lists are important in judging proper use, this
makes good sense. How could that concept be built into a meta database
of lists vs users? A simplistic approach might be to say,
"It's unlikely that more than three lists are related, so the sign-on
threshold should be four [in time period x]"?
Jeremy Blackman wrote:
> On Tue, 12 Oct 1999, Steve Bergeon wrote:
> > So the conclusion is that over one, ask the owner? I can
> > live with that.
> What I tend to notice is when someone subscribes to a bunch of UNRELATED
> mailing lists. If I see someone subscribe to listar-support, listar-dev,
> and listar-cvs, I probably won't think twice about it...when I pick up a
> new package, I'll often subscribe to both support and development lists,
> and a CVS notification list if there is one.
> If I see someone subscribe to every publically-viewable list on a machine
> at once? THAT raises alarm bells.