On Mon, Sep 12, 2005 at 03:44:04PM +0200, Brad Knowles wrote: > I believe that you are right, in that the higher the ratio of > publicly advertised NTP servers to general population, the higher the > relative number of people in that country who have the necessary > knowledge to maintain such a server. At least, I believe that there > is a high correlation between these two factors. > > However, I'm not sure that I understand how this relates to the > issue being discussed. Can you enlighten me?
I'm talking about education. Just the fact alone that we are talking about abuse, blacklisting, networking alone implicates that we do know a thing or two about our work. Now, if we do care about those members that donate their server to the pool (and I'm not saying that we don't allready), we might just aswell educate and simplify the whole process for them, en let them learn who to survive this often black form magic. > > If we are, and we feel that we need to give every part of the world that > > we can serve an equal good list of good ntp servers, we should not be > > looking at this from a national point of view, but more or less from a > > boundry less approach. > > Insofar as this statement goes, I generally agree with the > principle of going with boundaries specified at a much higher level > than individual countries. > But to do this and to make the effort worth-while, I think you'd > have to have reasonably intimate knowledge of the routing between the > various countries, and as has been said before, that kind of network > topological state can (and does) change on a moment-by-moment basis. Well, lets trust ntp in stateing the best ntp server in the list you provided it. Let's trust our transits and peers (our hosts) in providing us that with something equal to or way beyond reasonable. I am not talking about security or abuse here, so don't bite my head of for the usage of this dreaded word. This is still a best effort project, and that's only just because of the simple fact that we all aware of the amount of issues that can and will pin us down eventually. > I think we're better off just throwing all the European countries > (at least, the Western European countries) into the same pot, and not > try to second-guess the local administrator any more than that. > West and east Europe, that would make sense to me. Let's not forget that statistics and graphs are still very important for the future. > > Now.. if we would realy like to create something that belongs to the > > best what we can get in time serving there is, we should do it alongside > > the pool, provide it as a service for the pool members only to improve > > the quality. > > Now that is a very interesting concept. So, we'd be bringing our > own idea of "Private Stratum-1" services to the members of the pool > themselves, and then having them turn around and provide "Public > Stratum-2" services to the rest of the world. That is a very > interesting concept, indeed! > > That would satisfy the chimeheads in the group that want to tweak > the last nanosecond out of their clock, while avoiding unnecessary > complexity in providing "good enough" time services to J. Random > Internet User, but doing so in a much more robust and scalable > fashion. > > Cool! It might be great return to those who offer their resources to the pool besides all the information allready provided by Ask. Next step.. well, 2 steps. Is this something that just Brad and I like, and think that can work? Keep in mind, the "best of the best" is a tad to high to kick of with.. but it can be done, without an doubt. I wonder how much work such an undertaking might take, and if this would fit in the pool it self. Ask? _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
