Re: URGENT Re: Heads up: Long AS-sets announced in the next few days
[ris in CC field changed to ris-int to avoid replies being ticketized. Also CCing my co-authors in Rome] Daniel Karrenberg wrote:
If there is resistance like that, even uninformed resistance, we cannot push things like this.
I agree. Obviously the NCC is the least appropriate entity to push anything in the face of the community, since its charter is to do what the community wants it to do. And since it will be AS12654 doing the announcements, the buck stops here. I do think that Henk's message lacked finesse though. This is not me being overenthusiastic, this is a serious research project from a group with a proven record in the field of network discovery and a history of providing useful services to the community. Of course you can debate whether it is a good idea or not; every new technique has its fair share of detractors. And we wouldn't have gone ahead with the experiments in the face of such resistance either. But writing this off with a simple "On this occasion, he became a bit overenthusiastic though" is a little demeaning of the group's work. Anyway, with that behind us:
I beleive you should compose a very well drafted message explaining exactly
- what we are proposing to do - what effects we expect - why the effects people are concerned about will not occur
and ask for comments. This message should be OKed by Henk and maybe it would be good to have it signed by both Henk and Lorenzo. We should do this relatively quickly but without loosing the "well drafted".
I would be happy to do this if you think it stands a chance. The issue I fear is that people in favour or neutral to these experiments won't reply to the thread and only a (possibly small number of) more conservative and/or uninformed people will be quick to criticize and opt-out. How much detail do you think we need? Do you think that to convince people we need to describe all our algorithms (some of which we haven't even fully developed yet), or do you think a more general description would be enough? Of course we would still provide much more detail on the principles we want to use and how we are trying to accomplish it. Cheers, Lorenzo
At 00:55 04/03/2005, Lorenzo Colitti wrote:
But writing this off with a simple "On this occasion, he became a bit overenthusiastic though" is a little demeaning of the group's work.
It was not intended as such, this is indeed serious research by people who know what they are doing. Where you became overenthousiastic was when you wanted to experiment with other people's resources without their explicit approval.
The issue I fear is that people in favour or neutral to these experiments won't reply to the thread and only a (possibly small number of) more conservative and/or uninformed people will be quick to criticize and opt-out.
That's why I suggested opt-out rather than opt-in. We tell people what we want to do and why it won't have any negative effects. If somebody is still not convinced, he can opt-out. The people neutral or in favour do not have to do anything. Henk ------------------------------------------------------------------------------ Henk Uijterwaal Email: henk.uijterwaal(at)ripe.net RIPE Network Coordination Centre http://www.amsterdamned.org/~henk P.O.Box 10096 Singel 258 Phone: +31.20.5354414 1001 EB Amsterdam 1016 AB Amsterdam Fax: +31.20.5354445 The Netherlands The Netherlands Mobile: +31.6.55861746 ------------------------------------------------------------------------------ Look here junior, don't you be so happy. And for Heaven's sake, don't you be so sad. (Tom Verlaine)
On 04.03 00:55, Lorenzo Colitti wrote:
... I do think that Henk's message lacked finesse though.
To the contrary; I consider it very appropriate and well formulated.
I beleive you should compose a very well drafted message explaining exactly
- what we are proposing to do - what effects we expect - why the effects people are concerned about will not occur
and ask for comments. This message should be OKed by Henk and maybe it would be good to have it signed by both Henk and Lorenzo. We should do this relatively quickly but without loosing the "well drafted".
I would be happy to do this if you think it stands a chance.
The issue I fear is that people in favour or neutral to these experiments won't reply to the thread and only a (possibly small number of) more conservative and/or uninformed people will be quick to criticize and opt-out.
At the moment the nay-sayers have it clearly because we proposed to use their resources without their consent.
How much detail do you think we need? Do you think that to convince people we need to describe all our algorithms (some of which we haven't even fully developed yet), or do you think a more general description would be enough?
Exactly what I wrote: """""" - general description of the routes announced - prefixes - as paths / as sets - what do we expect to see in general terms - routing table effects - forwarding table effects - do they matter with these prefixes ? no! - concerns we have heared - c1; why it is not an issue - c2; why it is not an issue - ... Are there any other concerns? """"""""
Of course we would still provide much more detail on the principles we want to use and how we are trying to accomplish it.
That's optional but some of it may help to convince people. Daniel
participants (3)
-
Daniel Karrenberg -
Henk Uijterwaal -
Lorenzo Colitti