> 1. It would be nice to be able to have a utility to ping the server of the
> user, regardless of what it is. This would assist the average channel op in
> determining if there is a *network* lag that is causing a user to be slow in
> responding, or if the user has turned off CTCP replies.
We'll have to think about this one, but it might be possible.
> We would not need to know what server it was I think, but just how far away
> it was lag wise. (since it would always be a public server anyway, you might
> provide the server name too ?? That would be *nice* to know sometimes too,
> but opens up the user for some attacks (??), so may be better to keep it
> hidden? You guys know this stuff far better than us ;) )
Server names are hidden across the board; that is, you can find out what
server you're on with a /whois, but you cannot find out what server another
person is on from the /whois output, regardless of if they're on your
server or not (which is why we have to think about the "ping a user's
server" suggestion).
> 2. It would be nice to be able to see all of the *regular* servers on our
> side of the split. This information would be invaluable in determining
> where a lost user/channel bot might be found.
Again, we're hiding all server names, providing the appearance of a cloud
structure to users. Hopefully, we'll eventually move to make it a cloud
structure in truth, though that might be a few years.
> On a related note, I noticed that nick collision messages are still showing
> some server/routing information:
>
> *** Chrilith (~[EMAIL PROTECTED]) Quit (Killed
> (*.undernet.org (NewYork-R.NY.US.Undernet.Org <-
> arlington.va.us.undernet.org (nick collision from same user@host))))
>
> (Shown on Amsterdam 22 AUG 01 )
The kill reason is supplied by the killing server; pl14 and prior do not
properly hide this information. This means that the server that generated
the kill is still at pl14 or below. When the entire net has been upgraded,
you will cease to see kills like this.
> I hope I have not come accross as a complete nut. :)
Nope, not at all.
--
Kevin L. Mitchell <[EMAIL PROTECTED]>
