Re: Reducing nick-collision takeovers

[kl@xxxxxxxxxxx (Kaspar Landsberg)]

Hi,

you (Christophe Kalt) wrote:

| how often do this occur? (both ND and CD have timed out, it
| seems)

I have no idea. But obviously often enough since there passes hardly any  
time without someone joining for example #eu-opers asking for help because  
his/her channel has been taken over (of course, this doesn't mean that  
every takeover was done by nick collisions).

| are splits commonly long enough that ND and CD both time
| out?

Not all of them. But i suppose there are some servers which colliders  
chose as default server...

| on small networks, splits shouldn't be long (and the default
| delay is fine),

Agreed.

| on larger networks, splits can be longer.  It's easy to
| deal wit this administratively.  track the servers commonly
| used by abusers and deal with them (raising the delay is one
| option).

Yes, i didn't think of that. But then, this has to be done more actively  
(tracking the servers and mailing its admins) than currently.

| Do you realize that this is TS under disguise?

No, i didn't even think of TS. :) But you seem to be right somehow.

| fnc requires major change(s) to the protocol, and isn't
| possible to easily implement gradually.
| of course it's possible, but on top of that, fnc is likely
| to be disruptive.

A conjunction of {ND,CD} and FNC would certainly be less disruptive since  
{ND,CD} are reducing a lot the number of collisions. The great advantage  
would still be that nick collisions (resulting in kills) would almost not  
happen anymore... (which is not proved tho because (afaik) fnc has not yet  
been tested on a larger net)

Bye, Kasi

-- 
Kaspar Landsberg, <kl@xxxxxxxxxxx>