On Thu, Sep 14, 2006 at 09:26:46PM +0200, Robert Millan wrote: > > If we're going to try to do smart detection of a preferred mirror, it seems > > to me that geopolitical boundaries are not the most significant factor. > > Intervening network topography, available bandwidth, and (in some cases, > > like with most ISPs in Australia) off-net traffic costs play a much bigger > > role in determining a user's mirror presence, IME. So if we're going to go > > down this road, wouldn't it be better to design for all of this?
> Uhm right. So iplocation really only makes sense for localechooser, but it's > of > no use there since we don't have network yet. Maybe this is a dead end. > As for mirror selection, you mean something like netselect? Yes, something /like/ netselect, though I had in mind a solution that would query a centralized server for the "closest" mirror from a set based on BGP tables, removing the need for time-consuming pings of individual mirrors :) -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. [EMAIL PROTECTED] http://www.debian.org/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
