Re: Interim meetings planning [was Softwires Interim Meeting]
Hi John, Thanks for your input. See below, in-line. Regards, Jordi De: John C Klensin [EMAIL PROTECTED] Responder a: [EMAIL PROTECTED] Fecha: Thu, 26 Jan 2006 16:53:51 -0500 Para: [EMAIL PROTECTED] CC: ietf@ietf.org ietf@ietf.org Asunto: Re: Interim meetings planning [was Softwires Interim Meeting] Jordi, Let me make a very general observation, based on my experience with the IETF. Where administrative procedures are concerned, the IETF functions well when the IESG is given general guidance by the community but then applies good judgment and discretion to the situations that arise. If the community observes that the IESG's judgment is not good enough, we readjust the general guidance -- by complaints to individual ADs, by notes to the IESG, by discussion on the IETF list, or even via appeals if that is needed. If one or more specific ADs regularly abuse that discretionary authority, they get abused in return on mailing lists, in plenaries, in discussions with nomcoms, and, if necessary, by recalls or at least serious discussions of recalls. Agree, but I feel that there is a lack of administrative procedure here, which already caused some troubles. I'm trying to avoid repeating mistakes, improving the guidance and making sure the process is more open and fair. I fear that in this kind of situation (an Interim meeting unfairly setup), an appeal will not sort out the problem. By contrast, when we try to write down the specific rules that should be applied, or the list of rules or steps they should use in making decisions, and try to reach consensus on them, two things happen, and happen over and over again: (1) We spend (I would say waste, but you may reasonably disagree) a huge amount of time discussing both the details of the proposals and whether or not the proposals are appropriate. In my personal opinion, the IETF would be a better and more effective place to get work done if a higher percentage of the community's energy went into technical work than into discussions of procedural details. Agree in that we need more people, and working more in technical documents, but we like it or not, procedures are needed to make sure that the technical work can be followed up. Otherwise, we may break our main rule: consensus, not sure if say here also fairness. (2) We get the details wrong, resulting in more time being spent correcting the details of the rules that, IMO, we might have been better off not having in the first place. I really much prefer to avoid problems when they may be too late to get recovered. Details need to be put in place for that. In addition, if we get the details seriously wrong, the IESG has historically responded by applying a meta-rule. That meta-rule is that their first obligation is to keep the IETF functioning. And, on that basis, once they conclude that whatever has been written down and established by consensus makes no sense, they simply ignore ignore it. That action creates a very unhappy and unhealthy environment, whether it is appealed or not, but often may still be better than the alternative. Not really sure if I've seen a case like the one that you describe, but really doubt it may be the case for what I'm proposing. So, for a case like this, I suggest that you might want to have a conversation, with any AD you think is likely to listen, about whether the present guidance is sufficient and whether additional guidance or discussion would help. ADs persuading other ADs about IESG matters and customs can be far more effective than any of us writing I-Ds to pin down details. If there are no ADs with whom you feel that you can have such a conversation, then I think you need to chat with the Nomcom but I, at least, have generally found ADs to be receptive on these sorts of issues... even ADs with whom I generally don't get along well (not that there are any of those on the current IESG). Agree, and indeed my proposal was the outcome of a few email exchange with an AD ;-) He actually suggested to include this in the venue-selection-criteria document, which for me, and also talking to others, was not the best choice ... Let's try to avoid yet another round of stretched-out discussions on, e.g., how many days constitutes adequate notice, how to measure the accessibility of a location, what conditions are sufficient to justify an exception, and whatever other questions start to arise as soon as one tries to lock down specific rules. If an interim meeting is announced that is likely to cause specific harm (as distinct from being an issue of principles about dates and locations), complain to the relevant AD and, if necessary, appeal the decision to approve the meeting. But most of us, yourself certainly included, have more productive things to do with our time than starting another one of these procedural debate threads. Just my opinion, of course. And thanks a lot
Interim meetings planning [was Softwires Interim Meeting]
Hi Harald, In my opinion the 30 days rule is a good one, it may be possible to make it a bit flexible, just indicating 3-4 weeks before a meeting instead of 30 days. My comment, based on very recent experience, is that the rest of the Interim meeting planning procedure must be described more explicitly. The idea is not to have this in a new draft, or if actually we want it, make it in specific one, not mixed with anything else. The actual rules on this are at http://www.ietf.org/IESG/STATEMENTS/Interim-meetings.txt The area directors will evaluate proposed interim meetings and conference calls to be sure that that the location, timing, etc. do not unfairly favor some subset of the potential attendees and that the proposed meeting or call will not unfairly bias the working group discussions. This is my proposal: Interim meetings must be planned ahead, in order to facilitate a broader participation, reduce clashes with other meetings or events, and bring the cost down for the attendees. The need for them is typically recognized during a main IETF meeting. In order to allow a fair participation, they need to be announced in the relevant WG mail exploder 60 days in advance of the expected meeting date with a request for candidate hosts and dates. Typically this means announcing it already during the running main IETF meeting which recognized the need for that. This will also facilitate the planning for some of the participants. Once the intention to hold the meeting has been announced, in one week time, the hosting offers should be confirmed to the relevant WG mail exploder. The co-chair and ADs will provide a summary at the end of that week, excluding possible identified clashes and the WG will have one more week for the a vote on that. All the process must be made openly in the mail exploder in order to allow certain interaction for possible small variations during the 1 week proposal time. The final decision should be taken at least 6 weeks in advance to the meeting. The ADs will base its decision in the mailing list vote, prioritizing the participation of those which have work related to the meeting target and indicate their interest in attending. In any case the target is to not unfairly favor some subset of other potential attendees in such way that the proposed meeting will not unfairly bias the working group discussions. The ADs should consider not holding the meeting if there is an impossibility for key contributors with existing work to attend. Is expected that the meeting is organized 3-4 weeks ahead of a full IETF meeting, unless the AD says otherwise. This could be represented as: 60(+) days - Announcement of Intention to hold it 3-4 weeks /--\/\ IETF Int. m. announced Int. m. IETF |---|-|||-| | 1 week| 1 week 6(+) weeks | | | +-- start of voting 1 week period + start of proposals 1 week period How do you feel about this ? Regards, Jordi PS: May be we need something similar for conference calls, but the timing should be different De: Harald Tveit Alvestrand [EMAIL PROTECTED] Responder a: [EMAIL PROTECTED] Fecha: Tue, 24 Jan 2006 15:15:04 -0800 Para: [EMAIL PROTECTED], ietf@ietf.org ietf@ietf.org Asunto: Re: Softwires Interim Meeting --On 24. januar 2006 18:08 -0400 JORDI PALET MARTINEZ [EMAIL PROTECTED] wrote: In order to avoid this happening again, I'm working in some more clear suggestions for rules on how to adequately plan Interim meetings. I will circulate them ASAP. I wonder if that's the right approach. the original impetus for the 30 day rule was a situation where we had a non-US IETF (I think it was Adelaide), and suddenly were hit by a flurry of requests for interim meetings a week or two before or afte the IETF meeting - all of them intending to be in the US. The interpretation of some was that this looked like an attempt by US participants to avoid the expense of going overseas, leading to a perception that they thought it was fair that overseas participants always paid the cost of participating in US meetings, but not vice versa. In this case, the IETF meeting is in the US, and the interim meeting is not - so the foot may be in the other mouth, as the saying goes. If making rules, I'd say 30 days is the norm. It's a rule unless the AD says otherwise; the AD's decision has to be published (so that we can see who to blame if the community thinks it's not OK. But be careful what's a rule and what's advice, and don't mix too many topics into one document. it destroys the ability to get finished. Harald ___ Ietf mailing list Ietf@ietf.org
Re: Interim meetings planning [was Softwires Interim Meeting]
Jordi, Let me make a very general observation, based on my experience with the IETF. Where administrative procedures are concerned, the IETF functions well when the IESG is given general guidance by the community but then applies good judgment and discretion to the situations that arise. If the community observes that the IESG's judgment is not good enough, we readjust the general guidance -- by complaints to individual ADs, by notes to the IESG, by discussion on the IETF list, or even via appeals if that is needed. If one or more specific ADs regularly abuse that discretionary authority, they get abused in return on mailing lists, in plenaries, in discussions with nomcoms, and, if necessary, by recalls or at least serious discussions of recalls. By contrast, when we try to write down the specific rules that should be applied, or the list of rules or steps they should use in making decisions, and try to reach consensus on them, two things happen, and happen over and over again: (1) We spend (I would say waste, but you may reasonably disagree) a huge amount of time discussing both the details of the proposals and whether or not the proposals are appropriate. In my personal opinion, the IETF would be a better and more effective place to get work done if a higher percentage of the community's energy went into technical work than into discussions of procedural details. (2) We get the details wrong, resulting in more time being spent correcting the details of the rules that, IMO, we might have been better off not having in the first place. In addition, if we get the details seriously wrong, the IESG has historically responded by applying a meta-rule. That meta-rule is that their first obligation is to keep the IETF functioning. And, on that basis, once they conclude that whatever has been written down and established by consensus makes no sense, they simply ignore ignore it. That action creates a very unhappy and unhealthy environment, whether it is appealed or not, but often may still be better than the alternative. So, for a case like this, I suggest that you might want to have a conversation, with any AD you think is likely to listen, about whether the present guidance is sufficient and whether additional guidance or discussion would help. ADs persuading other ADs about IESG matters and customs can be far more effective than any of us writing I-Ds to pin down details. If there are no ADs with whom you feel that you can have such a conversation, then I think you need to chat with the Nomcom but I, at least, have generally found ADs to be receptive on these sorts of issues... even ADs with whom I generally don't get along well (not that there are any of those on the current IESG). Let's try to avoid yet another round of stretched-out discussions on, e.g., how many days constitutes adequate notice, how to measure the accessibility of a location, what conditions are sufficient to justify an exception, and whatever other questions start to arise as soon as one tries to lock down specific rules. If an interim meeting is announced that is likely to cause specific harm (as distinct from being an issue of principles about dates and locations), complain to the relevant AD and, if necessary, appeal the decision to approve the meeting. But most of us, yourself certainly included, have more productive things to do with our time than starting another one of these procedural debate threads. Just my opinion, of course. john --On Thursday, 26 January, 2006 10:02 -0400 JORDI PALET MARTINEZ [EMAIL PROTECTED] wrote: Hi Harald, In my opinion the 30 days rule is a good one, it may be possible to make it a bit flexible, just indicating 3-4 weeks before a meeting instead of 30 days. My comment, based on very recent experience, is that the rest of the Interim meeting planning procedure must be described more explicitly. The idea is not to have this in a new draft, or if actually we want it, make it in specific one, not mixed with anything else. The actual rules on this are at http://www.ietf.org/IESG/STATEMENTS/Interim-meetings.txt The area directors will evaluate proposed interim meetings and conference calls to be sure that that the location, timing, etc. do not unfairly favor some subset of the ... ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
