Re: 6bone space used still in the free (www.ietf.org over IPv6 broken) (Was: why same names, was Re: NANOG 40 agenda posted)
The IPv6 connectivity problems, at least the ones that I and some others encountered, where resolved yesterday. Thanks to all the folks involved who made that happen! Mike Leber wrote: [..] > Would you similarly disconnect a nonresponsive customer because they used > a /30 from RFC1918 space on a point to point link with you? That link should not have existed in the first place. Also RPF filters would catch any packets being sourced from that block. [..] >> Of course, Neustar, who are hosting www.ietf.org, might also want to >> look for a couple of extra transit providers who can provide them with >> real connectivity to the rest of the world. > > That won't renumber Bill Manning's links if that is the problem you are > trying to fix. Not much to fix when the person in question doesn't want to fix it. That comment was more meant to point that Neustar should have multiple upstreams. [..] > BTW, Jeroen does have a valid beef, [EMAIL PROTECTED] used to not be > handled by our normal engineering staff. It was somebody's part > time side project. This has changed with the migration of our IPv6 > network into our core. Since IPv6 is available via all core routers > for customers on the same links as their IPv4 connection, all > Hurricane network engineers are now required to be able take care > of issues with it. That is great to hear, keep the good work coming! Greets, Jeroen signature.asc Description: OpenPGP digital signature ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
RE: Putting IPR on the IETF consensus proccess (was: beware of fake pills)
Absolutely, I have had an attempt to set up a WG explode over the difference. It would be good to standardize the license terms. What I would like to see is a Microsoft or an IBM offer the use of existing terms such as they have already written under a creative commons license. In other words, no discussion, no interest group, no process, just a 'make use of this if you like'. We don't all need to use the same set of terms. It is unnecessary for us all to specify the same objectives in different terms. Lets push the whole set of IPR issues out to the folk in the creative commons community who live for such stuff. I would like to be able to call a pre-standards workshop under IPR terms splunge and for everyone attending to clearly understand the consequences, obligations etc. that splunge implies and not have to spend a week educating their corporate lawyers to let them attend. > -Original Message- > From: Bernard Aboba [mailto:[EMAIL PROTECTED] > Sent: Friday, May 25, 2007 11:51 PM > To: ietf@ietf.org > Subject: Re: Putting IPR on the IETF consensus proccess (was: > beware of fake pills) > > With all due respect, broad defensive non-assert clauses are > quite different from RF licenses. For an analysis of the > differences, see the article below: > > http://www.law.uchicago.edu/files/lichtman/def-susp.pdf > > Brian Carpenter said: > > It's a defensive non-assert disclosure, which IMHO is > equivalent to RF for anyone who plays nicely. Actually a > defensive non-assert may indirectly *protect* a normal > implementor, when you think about its impact on a third party > implementor who does try to assert a patent. > > ___ > Ietf mailing list > Ietf@ietf.org > https://www1.ietf.org/mailman/listinfo/ietf > ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
RE: On the IETF Consensus process
I think we need an explicit prohibition on a member of the IESG being a Working Group chair. This does not happen often but it has done in the past and it has caused a real problem. Simply prohibiting an AD being chair of a WG in their own area is not enough. The problem with consensus is how you decide to count the undecideds/neutrals. In most cases of controversy there will be a small group pro, a small group con and the bulk of the WG will be somewhere inbetween. If the breakdown is 25%/25%/50% a biased chair can effectively decide the outcome by choosing to interpret 'no objection' as 'no support' or vice versa. Consensus is always the preferred basis for decision and usually the best basis. There are situations where putting 'consensus' above pragmatics leads to a proposal that is never going to be deployed. If we are proposing to change the internet infrastructure and do not have the support of the ISPs or the core DNS providers or whatever key constituency may be required to put a proposal into effect then we need to have a mechanism for dealing with that situation. Another limitation to consensus is where there is a conflict between two groups. There may be 100% consensus in group A that approach 1 is best, but serious objections from group B. Or as frequently happens there may be a dispute in group A as to the likelihood of objection a proposal will receive from group B. The appeals process is not optimal here because it only begins after the process failure has occurred. The recall process is unworkable and unless someone screws up really bad really early is not going to meaningfully shorten a tenure that would otherwise expire anyway. > -Original Message- > From: Brian E Carpenter [mailto:[EMAIL PROTECTED] > Sent: Sunday, May 27, 2007 10:57 AM > To: Lakshminath Dondeti > Cc: John C Klensin; IETF Discussion; Jeffrey Hutzelman > Subject: Re: On the IETF Consensus process > > On 2007-05-26 00:46, Lakshminath Dondeti wrote: > ... > > > Coming to the fear of retaliation issue (I guess we are > talking about > > unethical behavior on part of an IESG or IAB member at this > point), we > > need to find a way to fix that. On this, we seem to be > worse off than > > the outside world. Let's take the US as an example (or we can take > > India too; I am sorry, but those are the only two countries I am > > somewhat familiar with): there are always checks and balances. > > That's exactly why we have the appeals process, the recall > mechanism, and an independent Nomcom. And speaking for myself > (and probably > others) that's why more transparency is always good, as long > as it doesn't prevent the IAB and IESG being able to discuss frankly. > > If you can suggest additional checks and balances, please do! > > Brian > > ___ > Ietf mailing list > Ietf@ietf.org > https://www1.ietf.org/mailman/listinfo/ietf > ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
consensus and anonymity
Hallam-Baker, Phillip wrote: The problem with consensus is how you decide to count the undecideds/neutrals. In most cases of controversy there will be a small group pro, a small group con and the bulk of the WG will be somewhere inbetween. If the breakdown is 25%/25%/50% a biased chair can effectively decide the outcome by choosing to interpret 'no objection' as 'no support' or vice versa. One thing that occurs to me is that there is usually a huge disconnect between the participation in hums at a meeting and the email equivalent on the working group list. I'd say that it's typically between one and two orders of magnitude at a meeting more hands/hums than on the list. And of course, on the list it's usually just a rehash of the same active participants with a few stragglers thrown in. Maybe part of the problem with the "official" consensus taking on the list is that it isn't sufficiently anonymous? It's pretty easy in a crowd to hum or put up your hand in a sea of others; on the list, it requires quite a bit more conviction. Apathy is the other likely reason, but there's not much we can do about that short of working group demolition derby videos or suchlike. So might having the ability to contact the chairs in private to register their preference be reasonable? I don't recall seeing this in any of the working groups I've participated in. Mike ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
RE: On the IETF Consensus process
--On Thursday, 31 May, 2007 09:00 -0700 "Hallam-Baker, Phillip" <[EMAIL PROTECTED]> wrote: > I think we need an explicit prohibition on a member of the > IESG being a Working Group chair. This does not happen often > but it has done in the past and it has caused a real problem. > Simply prohibiting an AD being chair of a WG in their own area > is not enough. Phillip, I think this shows a conflict between two principles. I would like to see if we can thread it with some sophistication, rather than making more and more rigid rules. I'd like to think that the majority of our WGs are relatively non-controversial and that the key consideration in the choice of Chairs is simply what gets the work done most efficiently. So, to take the most obvious case, if there is a WG that is making reasonable progress and perhaps is getting close to producing products, has not been controversial, but has been chaired or co-chaired by someone who then becomes an AD, I'd rather ask for good judgment than for an immediate resignation. The decisions and constraints facing the Nomcom are hard enough without their having to try to evaluate what the loss of a chair would do to an important WG. At the other extreme, I suggest that having an AD as chair of a high-controversy WG, even if it is in a different area, represents seriously bad judgment for the reasons you suggest and others. Personally, I think the bad judgment would be severe enough that they ought to be calls for resignation (as an AD, not as WG Chair). I also believe that, if such informal calls were ignored, recalls would be in order. Not only does a little bit more flexibility about this -- coupled with the community insisting on good judgment-- make more sense to me than a firm rule, but there is a scenario you didn't outline that would make a firm rule less useful than I think you are suggesting. Suppose someone had been chairing a WG for some time. She has gotten technically and emotionally invested in the particular decisions the WG has been making. The WG has work that is late in the process -- nearing or perhaps already in, Last Call. Then she gets appointed AD, takes that position, and resigns as WG Chair as your proposed rule would require. I think that, in most cases, the resignation would accomplish nothing as a "cure" for any possible bias. Whether the individual continues as Chair or not, I think the important thing is that the situation, and the way in which it was going to be handled within the IESG, be handled in a fashion that was open enough, and discussed openly enough with relevant people in the community (including dissenters within the WG), that the community could evaluate whether or not the basic integrity of the process was being upheld.As above, if so, fine. If not, we would need appeals and resignations or recalls, not narrowly-written rules about positions and resignations. john ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
--On Thursday, 31 May, 2007 10:22 -0700 Michael Thomas <[EMAIL PROTECTED]> wrote: > So might having the ability to contact the chairs in private > to register their > preference be reasonable? I don't recall seeing this in any of > the working groups I've participated in. But I certainly know of cases where it has been done, often with copies to the relevant AD as a safety precaution. It is worth noting that even the new instructions about last calls indicate the option of sending comments only to the IESG, rather than the IETF list, if there are exceptional circumstances. Judgment about what constitutes the latter are apparently up to the submitter. I think this is more evidence that we need more flexibility and good sense, not more rigid rules. john ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: [Tsvwg] Re: Last Call: draft-ietf-tsvwg-cc-alt (SpecifyingNewCongestion Control Algorithms) to BCP
Mitchell- > The SHOULD type terminology is > what I thought is standard. This is not a protocol spec. Such language is a little fuzzy in documents that are not specs. Unless folks loudly and broadly complain it seems to me that the language in the current document has been agreed to. > And why wouldn't a minimum timeframe be specified? And items > listed that state how an implementation exits this experimental > phase, versus subjective evaluation. Ie, a vote of a working > group that requires xx% acceptance. We don't vote. It happens to be part of our charm. :) I personally don't see why we'd place a time frame on things. It seems more important to look at the available evidence and experimental results than to try to pick some arbitrary amount of time. Again, unless others start yelling, I think we have consensus on this document. > Secondly, I have never understood the non-requirment > of a REFERENCE implementation. The Reference implmentation > IMO, doesn't have to be source. I am not sure how an "reference implementation" does not "have to be source". But, we're talking about algorithms here. It seems clear that these algorithms will have to be well-specified or the WG (whatever WG that happens to be) is not going to consider the spec. > Lastly, should a accepted CC algor ever be re-reviewed >to see if it has aged well??? And needs updating. I think that any time folks have an update they should bring it up. I would consider that what has actually happened in practice. > > > 4) Support for a CC algorithm in one environment, ie LFN.. > > > > I don't understand what you are getting at here. But, see guideline > > (2). > > Can a implimentation be offered to work in ONLY 1 environment? See guideline (8): As a similar concern, if the alternate congestion control mechanism is intended only for specific environments (and not the global Internet), the proposal should consider how this intention is to be carried out. The community will have to address the question of whether the scope can be enforced by simply stating the restrictions or whether additional protocol mechanisms are required to enforce the scoping. The answer will necessarily depend on the change being proposed. (Which is slightly tweaked wording from the current version in the repository (in response to an IESG comment), but the intent is the same.) > All implementations that I have worked on, have a stated > set of configurables with default values. If a well known accepted > implimentation is significantly modified, at what point is it > considered a new implementation. Ie: Removing fast acks in > small in-flight segment environments I don't think I can give you a hard-and-fast rule. This is just something the WG would have to work through. allman pgpTo7HhxvEgb.pgp Description: PGP signature ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
On 5/31/07 2:49 PM, "John C Klensin" <[EMAIL PROTECTED]> wrote: > I think this is more evidence that we need more flexibility and > good sense, not more rigid rules. Well, what's under description really isn't consensus decision-making processes - what's being argued is a sort of voting. Rather than getting stuck on definitions, though, I think it's probably worthwhile to frame the discussion in terms of intentions - what the original intent was, what the current intent is, whether or not there's a mismatch, and how to get the process to look like what's intended, if it currently doesn't (and wow, do I think it does not). Arguing over percentages doesn't, I think, go very far towards figuring those things out. I think I understand what used to be intended but I'm not sure I can articulate it, but I do think that the problems around decision-making come down to a few things: 1) not-that-great decision facilitation skills on the part of some chairs and leaders; 2) too many participants who'd be happy shutting down the whole process rather than accomodating a decision they don't agree with 3) the organization is just too large for a touchy- feely decision-making style. But it seems to me that if you're going to go with some form of voting, and that's what's happening in practice, it would be better to design a fair voting process. If you're going try to do consensus better you have to figure out how to deal with the kvetches in a way that doesn't shut out dissent while still allowing decision-making to move forward. Either one is hard to do but I'm not sure that doing nothing is an option. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Excellent point about the disconnect between meeting room hums and opinions on the lists. But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. The IETF should be more transparent and allow at least a distributed verification process and not a centralized hierarchical process. Lakshminath On 5/31/2007 10:22 AM, Michael Thomas wrote: Hallam-Baker, Phillip wrote: The problem with consensus is how you decide to count the undecideds/neutrals. In most cases of controversy there will be a small group pro, a small group con and the bulk of the WG will be somewhere inbetween. If the breakdown is 25%/25%/50% a biased chair can effectively decide the outcome by choosing to interpret 'no objection' as 'no support' or vice versa. One thing that occurs to me is that there is usually a huge disconnect between the participation in hums at a meeting and the email equivalent on the working group list. I'd say that it's typically between one and two orders of magnitude at a meeting more hands/hums than on the list. And of course, on the list it's usually just a rehash of the same active participants with a few stragglers thrown in. Maybe part of the problem with the "official" consensus taking on the list is that it isn't sufficiently anonymous? It's pretty easy in a crowd to hum or put up your hand in a sea of others; on the list, it requires quite a bit more conviction. Apathy is the other likely reason, but there's not much we can do about that short of working group demolition derby videos or suchlike. So might having the ability to contact the chairs in private to register their preference be reasonable? I don't recall seeing this in any of the working groups I've participated in. Mike ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
RE: [Hipsec] Re: Last Call: draft-ietf-hip-applications (Using the HostIdentity Protocol with Legacy Applications) to Experimental RFC
Pekka, Thanks for your review of this draft. I'm cc'ing ietf@ietf.org to avoid cross-posting, and will summarize later to hipwg list. Responses inline below. > -Original Message- > From: Pekka Savola [mailto:[EMAIL PROTECTED] > Sent: Wednesday, May 30, 2007 12:25 AM > To: ietf@ietf.org > Cc: [EMAIL PROTECTED] > Subject: [Hipsec] Re: Last Call: draft-ietf-hip-applications > (Using the HostIdentity Protocol with Legacy Applications) to > Experimental RFC > > On Mon, 14 May 2007, The IESG wrote: > > The IESG has received a request from the Host Identity > Protocol WG (hip) > > to consider the following document: > > > > - 'Using the Host Identity Protocol with Legacy Applications ' > >as an Experimental RFC > > > > The IESG plans to make a decision in the next few weeks, > and solicits > > final comments on this action. Please send substantive > comments to the > > ietf@ietf.org mailing lists by 2007-05-28. Exceptionally, > > comments may be sent to [EMAIL PROTECTED] instead. In either > case, please > > retain the beginning of the Subject line to allow automated sorting. > > Some slightly late comments below. > > My main (meta-) issue with the draft is that it's not clear from how > the draft is written what is the goal of the draft. It seems it's > providing an informative overview of different approaches for HIP > experiments; it does not aim to provide any normative specification > for HIP implementations or applications, even to make the experiments > using different approaches to use more uniform methods. > > Is this correct? Yes. > If yes, maybe this could be made more > explicit, e.g., > by adding 'Overview on' at the start of the title and similar other > modifications. On the other hand, if this spec is intended > to be used > somehow by HIP or application implementations, I believe text would > likely need significant rework; there is very little explicit > guidance > or explicit specification as it is and very little text that would > help in creating interoperable implementations. I would not have a problem with renaming to something like "Overview of Approaches to Using the Host Identity Protocol with Legacy Applications". Do you think such a title would be better? The purpose of this document is to provide an overview of the various ways that HIP can be used without changing the applications, and the possible issues with the different approaches. It was not our goal to create specification for interoperable implementations, since the implementation issues are localized, and we have experience that different implementations choosing different solutions from these options can interoperate. Instead, we are trying to document some implementation experience to cover implementation issues not discussed in the other HIP drafts. > > substantial > --- > > While the text below concentrates on the use of the > sockets connect > system call, the same argument is also valid for other > system calls > using socket addresses. > > ==> I'm not sure if I will accept this assumption at its face value > without a reference. Are you sure all the socket-operating system > calls are basically equivalent? Has this been studied somewhere > (Miika/Mika's Master's thesis as one example?) more extensively? For > example, what does listen(LSI) or bind(LSI) mean? Section 3.4 seems > to discuss this a bit implying that all the socket system > calls aren't > necessarily similar. I am not sure that anyone has done an extensive review of all possible socket calls, but I think that this claim is valid in typical cases because, in practice, applications seem to work correctly. However, based on your comment, I propose to remove this sentence altogether because it should be clear from the later text that the discussion of connect() is intended as a typical example (e.g., section 3.1) and later in section 3.4 we discuss some caveats regarding wildcard values. > > Using DNS to map IP addresses to HIs: > >If the responder has host identifiers registered in the forward >DNS zone and has a PTR record in the reverse zone, the > initiating >system could perform a reverse+forward lookup to learn the HIT >associated with the address. [...] > > ==> does this cause a recursion problem with DNS resolver IP > addresses? I.e., you try > to look up HIP records by reverse+forward of node X by doing > queries to DNS > servers A and B, but end up querying DNS reverse+forward > records of A and B > through DNS first. I think this should work under normal > circumstances > but I can see some potential reliability issues; at least if > the DNS server > addresses are provisioned with HIP records but they don't support HIP, > you might end up hosing all your DNS lookups if the fallback > from trying HIP > to no-HIP isn't reliable. > > Is there already sufficient experience of these kind of > potential bootstrap >
Re: consensus and anonymity
On 5/31/07 3:21 PM, "todd glassey" <[EMAIL PROTECTED]> wrote: > i.e. interferring with others initiatives... this is a serious issue since > this is Tortuous Interferrence per se. Actually, Todd, what was I thinking of was people who aren't capable of graciously moving out of the way after they've made their point and "lost." People who threaten lawsuits and whatnot. They really make it extremely difficult to finalize a decision and they poison the decision-making process. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Just following up here... From: "Lakshminath Dondeti" <[EMAIL PROTECTED]> But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. We've been telling new WG chairs for several years that - they really need to have most discussions in public/on mailing lists, - we recognise that some people aren't comfortable challenging others in public, and - we recognise that this discomfort may be more common in some cultures than in others. So, for reasons that both John and Lakshminath identified, we've been asking WG chairs to encourage participants to engage in public discussions, but to be receptive to private requests for assistance on how to carry out those discussions. The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Possibly, but it is also possible that they do not beleive that the process was observed, they may disagree as to what the consensus was. If the wg chair is also on the iesg the appeals process is fatally compromised and littigation may be the only realistic prospect for redress. I don't think it is possible to predict which areas are going to be controversy free, some areas such as dns and routing may be intrinsically controversy prone but that is not the point. There needs to be an inflexible rule here. Flexibility is not advantageous. Sent from my GoodLink Wireless Handheld (www.good.com) -Original Message- From: Melinda Shore [mailto:[EMAIL PROTECTED] Sent: Thursday, May 31, 2007 12:26 PM Pacific Standard Time To: todd glassey; John C Klensin; Michael Thomas; Hallam-Baker, Phillip Cc: IETF Discussion; Jeffrey Hutzelman Subject:Re: consensus and anonymity On 5/31/07 3:21 PM, "todd glassey" <[EMAIL PROTECTED]> wrote: > i.e. interferring with others initiatives... this is a serious issue since > this is Tortuous Interferrence per se. Actually, Todd, what was I thinking of was people who aren't capable of graciously moving out of the way after they've made their point and "lost." People who threaten lawsuits and whatnot. They really make it extremely difficult to finalize a decision and they poison the decision-making process. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Anonymity means that a cabal can block progress without being held accountable. If you can't argue your case in public you should be asking why. Sent from my GoodLink Wireless Handheld (www.good.com) -Original Message- From: Lakshminath Dondeti [mailto:[EMAIL PROTECTED] Sent: Thursday, May 31, 2007 12:14 PM Pacific Standard Time To: Michael Thomas Cc: Hallam-Baker, Phillip; Brian E Carpenter; John C Klensin; IETF Discussion; Jeffrey Hutzelman Subject:Re: consensus and anonymity Excellent point about the disconnect between meeting room hums and opinions on the lists. But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. The IETF should be more transparent and allow at least a distributed verification process and not a centralized hierarchical process. Lakshminath On 5/31/2007 10:22 AM, Michael Thomas wrote: > Hallam-Baker, Phillip wrote: >> The problem with consensus is how you decide to count the >> undecideds/neutrals. In most cases of controversy there will be a >> small group pro, a small group con and the bulk of the WG will be >> somewhere inbetween. If the breakdown is 25%/25%/50% a biased chair >> can effectively decide the outcome by choosing to interpret 'no >> objection' as 'no support' or vice versa. >> > One thing that occurs to me is that there is usually a huge disconnect > between > the participation in hums at a meeting and the email equivalent on the > working > group list. I'd say that it's typically between one and two orders of > magnitude > at a meeting more hands/hums than on the list. And of course, on the > list it's > usually just a rehash of the same active participants with a few > stragglers thrown > in. > > Maybe part of the problem with the "official" consensus taking on the > list is > that it isn't sufficiently anonymous? It's pretty easy in a crowd to hum or > put up your hand in a sea of others; on the list, it requires quite a > bit more > conviction. Apathy is the other likely reason, but there's not much we can > do about that short of working group demolition derby videos or suchlike. > > So might having the ability to contact the chairs in private to register > their > preference be reasonable? I don't recall seeing this in any of the working > groups I've participated in. > > Mike > ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
From: "John C Klensin" <[EMAIL PROTECTED]> --On Thursday, 31 May, 2007 10:22 -0700 Michael Thomas <[EMAIL PROTECTED]> wrote: So might having the ability to contact the chairs in private to register their preference be reasonable? I don't recall seeing this in any of the working groups I've participated in. (Aside to Michael - this may mean that if there WERE private discussions, they were actually private! "if you HAD seen this..." :-) But I certainly know of cases where it has been done, often with copies to the relevant AD as a safety precaution. It is worth noting that even the new instructions about last calls indicate the option of sending comments only to the IESG, rather than the IETF list, if there are exceptional circumstances. Judgment about what constitutes the latter are apparently up to the submitter. It's worth mentioning that the recent change to last call instructions reflected a sense that back-channel discussions with the IESG should be the exception, not the rule. When the IESG discussed this, I asked about public last call comments that set off blizzards of postings, and multiple ADs thought transparency was worth having to dig out of a blizzard from time to time. I think this is more evidence that we need more flexibility and good sense, not more rigid rules. I rarely rephrase John's words, but in this case, I'm thinking "more good sense and flexibility", in that order :-) john Thanks, Spencer p.s. I also agree with John's point that changing WG chairs when one becomes an AD is not likely to improve quality or throughput in many cases, but I'm trying to rate-limit my NUMBER of postings these days... that's definitely the corner case for "no ADs/WG chairs at the same time". ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Oh, I understand cultural sensitivities, but I have never heard of not wanting to challenge in the public (except the disagreeing with the employer thing). The problem with that is that if people don't like something and can't speak up or will only speak through a chair or an AD, it allows natural avenues for abuse. The chair or the AD might as well be making decisions at that point. Even anonymous voting has verifiability as the crucial part of requirements. If our consensus process is not independently and openly verifiable, we might as well close shop! Lakshminath PS: BTW, I agree with Melinda that we should not allow a minority to block progress; in any type of consensus process, unfortunately some of us will be at the losing end of things. On 5/31/2007 12:33 PM, Spencer Dawkins wrote: The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
One thing that occurs to me is that in my initial message I implicitly felt that the room hands/hums were a more accurate assessment of consensus than the list. I guess that I should fess up that I've always felt that the "consensus is determined on the list" is something of a charming myth. Of course if we went with the room consensus, gaming the system would just be done by different means, so my feeling shouldn't be taken as endorsement of that as an alternative. Mike Lakshminath Dondeti wrote: Oh, I understand cultural sensitivities, but I have never heard of not wanting to challenge in the public (except the disagreeing with the employer thing). The problem with that is that if people don't like something and can't speak up or will only speak through a chair or an AD, it allows natural avenues for abuse. The chair or the AD might as well be making decisions at that point. Even anonymous voting has verifiability as the crucial part of requirements. If our consensus process is not independently and openly verifiable, we might as well close shop! Lakshminath PS: BTW, I agree with Melinda that we should not allow a minority to block progress; in any type of consensus process, unfortunately some of us will be at the losing end of things. On 5/31/2007 12:33 PM, Spencer Dawkins wrote: The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
> If our consensus process is not independently and openly verifiable, we > might as well close shop! a hum in a WG meeting is subject to the perceptions of the people in the room but its not clear that a fully verifiable process is needed since we specifically try to do rough consensus not majority vote - all we need to be able to verify is that "most" people support a path - if a proposal gets blocked by a secret whisper in the ear of the WG chair then things are very broken indeed if the result of a secret whisper results in the chair brings up topics for public discussion I'm not all that worried Scott ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Oh definitely. Gaming the system is easy with "consensus in the room" especially with hums and such. The IEEE has voting and people allegedly send zombies to raise hands. There are provisions to protect against that. Anyone can challenge and claim that there is block voting (people represent themselves and not their employers, just as in the IETF). The chair has a procedure to determine whether that is the case and consider all the votes from a single company as one vote. It's a good deterrent. The problem with consensus in the room is that it flies in the face of our mode of operation. If we have to really wait until a meeting to make decisions, we are introducing tremendous delays into the process; in that case, we should meet more often. Next, if consensus in the room is how we want to do it and we are resigned to the fact that some of us (even those in power misbehave), let's make it fair; let's declare that. I will then tell everyone who listens to send zombies to the IETF to hum/vote/whatever. I hope we haven't come to that! best, Lakshminath On 5/31/2007 1:08 PM, Michael Thomas wrote: One thing that occurs to me is that in my initial message I implicitly felt that the room hands/hums were a more accurate assessment of consensus than the list. I guess that I should fess up that I've always felt that the "consensus is determined on the list" is something of a charming myth. Of course if we went with the room consensus, gaming the system would just be done by different means, so my feeling shouldn't be taken as endorsement of that as an alternative. Mike Lakshminath Dondeti wrote: Oh, I understand cultural sensitivities, but I have never heard of not wanting to challenge in the public (except the disagreeing with the employer thing). The problem with that is that if people don't like something and can't speak up or will only speak through a chair or an AD, it allows natural avenues for abuse. The chair or the AD might as well be making decisions at that point. Even anonymous voting has verifiability as the crucial part of requirements. If our consensus process is not independently and openly verifiable, we might as well close shop! Lakshminath PS: BTW, I agree with Melinda that we should not allow a minority to block progress; in any type of consensus process, unfortunately some of us will be at the losing end of things. On 5/31/2007 12:33 PM, Spencer Dawkins wrote: The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
--On Thursday, 31 May, 2007 14:33 -0500 Spencer Dawkins <[EMAIL PROTECTED]> wrote: > So, for reasons that both John and Lakshminath identified, > we've been asking WG chairs to encourage participants to > engage in public discussions, but to be receptive to private > requests for assistance on how to carry out those discussions. > > The alternative - a WG chair who tells the working group that > the apparent WG consensus on the mailing list is being > overruled because of anonymous objections that the WG chair > cannot share with the WG, or because of private objections > that the WG chair is "channeling" from a back room - would > make voting seem reasonable (or, to use Mark Allman's > characterization in another thread, "seem charming"). But we have mechanisms for dealing with those kinds of problems. As an example, I would assume that, if someone challenged an assertion of consensus based on undisclosed comments, the first thing a competent AD or IETF Chair would do would be to ask for names or other ways to validate the input. The comments don't need to be public in order to be verified and, in an operation like the IETF, there is (or ought to be) a difference between the expectation of not needing to be associated with a particular position in public and anonymity when making those comments. If a WG Chair makes a decision based on private input and cannot readily demonstrate that the input actually occurred in whatever numbers are implied (on appeal if necessary), then that WG Chair should be removed. If I were a participant in a controversial WG --and I don't buy the story that one cannot predict or diagnose controversy and hence must treat everything that way-- I'd expect to see relatively independent co-chairs and/or an AD who was clearly independent of the chair. I'd expect that, if I felt a need to make a comment "in private", I'd write up the comment and copy both co-chairs and the AD, not try to whisper into someone's ear in a dark hall somewhere. I think that is just good sense (and, yes, Spencer, "good sense" belongs in front of "flexibility" if one must impute an order. It is very clear that we have a system that is open to attack if there are sufficiently many parties with sufficient malice. Until and unless we start seeing lots of such attacks, it seems to me that expecting good behavior and having ways to detect and fix bad behavior should it occur is a much better path than trying to invent rules that, inevitably, will be addressing the last attack and not the next one and that will bog us down further. I do believe that, if a decision is made that claims to be based on consensus, but the consensus is not obvious from public comments, the person making that decision has some obligation to explain the decision and, clearly, "lots of people whispered to me" should not be accepted as an explanation.I also believe that, if that doesn't occur, appeals should be initiated and upheld. But I'm having trouble seeing a real, rather than theoretical, problem here that justifies new rules or procedures. john ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Spencer Dawkins wrote: Just following up here... From: "Lakshminath Dondeti" <[EMAIL PROTECTED]> But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. We've been telling new WG chairs for several years that - they really need to have most discussions in public/on mailing lists, - we recognise that some people aren't comfortable challenging others in public, and - we recognise that this discomfort may be more common in some cultures than in others. So, for reasons that both John and Lakshminath identified, we've been asking WG chairs to encourage participants to engage in public discussions, but to be receptive to private requests for assistance on how to carry out those discussions. The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). This is not an alternative. If you are not willing to make your technical objections to a technical specification publicly, then they cannot be part of the IETF decision-making process. What's to prevent a WG Chair from "padding" the anonymous "votes"? If 5 people in public (WG meeting or mailing list) are for some proposal, and the Chair says, "I heard from 6 people who are against this, but don't want their identities known, so the proposal is rejected." Not acceptable. Thanks, Spencer Andy ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Hi, John, From: "John C Klensin" <[EMAIL PROTECTED]> --On Thursday, 31 May, 2007 14:33 -0500 Spencer Dawkins <[EMAIL PROTECTED]> wrote: So, for reasons that both John and Lakshminath identified, we've been asking WG chairs to encourage participants to engage in public discussions, but to be receptive to private requests for assistance on how to carry out those discussions. The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). But we have mechanisms for dealing with those kinds of problems. Fully agree. I was thinking about preventing a problem, and you're correctly pointing out that we have tools to deal with the problem if we can't prevent it. Not that this sounds like fun, but you're quite correct. ... and continue fully agreeing, especially with It is very clear that we have a system that is open to attack if there are sufficiently many parties with sufficient malice. Until and unless we start seeing lots of such attacks, it seems to me that expecting good behavior and having ways to detect and fix bad behavior should it occur is a much better path than trying to invent rules that, inevitably, will be addressing the last attack and not the next one and that will bog us down further. I do believe that, if a decision is made that claims to be based on consensus, but the consensus is not obvious from public comments, the person making that decision has some obligation to explain the decision and, clearly, "lots of people whispered to me" should not be accepted as an explanation.I also believe that, if that doesn't occur, appeals should be initiated and upheld. But I'm having trouble seeing a real, rather than theoretical, problem here that justifies new rules or procedures. Thanks, Spencer ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Andy Bierman wrote: Spencer Dawkins wrote: Just following up here... From: "Lakshminath Dondeti" <[EMAIL PROTECTED]> But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. We've been telling new WG chairs for several years that - they really need to have most discussions in public/on mailing lists, - we recognise that some people aren't comfortable challenging others in public, and - we recognise that this discomfort may be more common in some cultures than in others. So, for reasons that both John and Lakshminath identified, we've been asking WG chairs to encourage participants to engage in public discussions, but to be receptive to private requests for assistance on how to carry out those discussions. The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). This is not an alternative. If you are not willing to make your technical objections to a technical specification publicly, then they cannot be part of the IETF decision-making process. If that's true, then why do we have hums at wg meetings at all? A hum doesn't give the reasoning; it's a binary quantity. What's to prevent a WG Chair from "padding" the anonymous "votes"? If 5 people in public (WG meeting or mailing list) are for some proposal, and the Chair says, "I heard from 6 people who are against this, but don't want their identities known, so the proposal is rejected." Not acceptable. Dishonesty by the management is a problem regardless of what system we have. Most wg's these days have two chairs, so collusion would need to be at least that deep, and probably require an AD to be on board too. If that really were the case, I doubt any system is likely to perform very well. But this cultural thing does bug me. It seems unsatisfying to me that our pat answer to cultural differences is "become more western". The language issue is already asking quite a lot of the rest of the world. Mike Thanks, Spencer Andy ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Michael Thomas wrote: Andy Bierman wrote: Spencer Dawkins wrote: Just following up here... From: "Lakshminath Dondeti" <[EMAIL PROTECTED]> But, I wonder why anonymity is an important requirement. The mailing list verification has at least two properties that are more important to the IETF: the archives provide for anyone to be able to verify the consensus independent of the IETF hierarchy (chairs, ADs and whoever); further the archives provide a means to verify the consistency of any IETF participant, chairs or ADs at any given moment, candidates for WG chair and I* positions, and anyone in general. We've been telling new WG chairs for several years that - they really need to have most discussions in public/on mailing lists, - we recognise that some people aren't comfortable challenging others in public, and - we recognise that this discomfort may be more common in some cultures than in others. So, for reasons that both John and Lakshminath identified, we've been asking WG chairs to encourage participants to engage in public discussions, but to be receptive to private requests for assistance on how to carry out those discussions. The alternative - a WG chair who tells the working group that the apparent WG consensus on the mailing list is being overruled because of anonymous objections that the WG chair cannot share with the WG, or because of private objections that the WG chair is "channeling" from a back room - would make voting seem reasonable (or, to use Mark Allman's characterization in another thread, "seem charming"). This is not an alternative. If you are not willing to make your technical objections to a technical specification publicly, then they cannot be part of the IETF decision-making process. If that's true, then why do we have hums at wg meetings at all? A hum doesn't give the reasoning; it's a binary quantity. I think the inability of the IETF to make decisions in an open, deterministic, and verifiable manner is a major flaw. It promotes indecision and inaction. What's to prevent a WG Chair from "padding" the anonymous "votes"? If 5 people in public (WG meeting or mailing list) are for some proposal, and the Chair says, "I heard from 6 people who are against this, but don't want their identities known, so the proposal is rejected." Not acceptable. Dishonesty by the management is a problem regardless of what system we have. Most wg's these days have two chairs, so collusion would need to be at least that deep, and probably require an AD to be on board too. If that really were the case, I doubt any system is likely to perform very well. Only transparency can prevent corruption. But this cultural thing does bug me. It seems unsatisfying to me that our pat answer to cultural differences is "become more western". The language issue is already asking quite a lot of the rest of the world. I don't see the cultural bias here. I see the bias in "English only", but these are public standards being developed. Everybody in the IETF should be able to read all of the comments made on a draft, not just a privileged elite. Mike Thanks, Spencer Andy Andy ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
> "Andy" == Andy Bierman <[EMAIL PROTECTED]> writes: Andy> I think the inability of the IETF to make decisions in an Andy> open, deterministic, and verifiable manner is a major flaw. I for one do not wish for deterministic decision making in the IETF. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
> "Andy" == Andy Bierman <[EMAIL PROTECTED]> writes: Andy> This is not an alternative. If you are not willing to make Andy> your technical objections to a technical specification Andy> publicly, then they cannot be part of the IETF Andy> decision-making process. At one level I agree here. Andy> What's to prevent a WG Chair from "padding" the anonymous Andy> "votes"? If 5 people in public (WG meeting or mailing list) Andy> are for some proposal, and the Chair says, "I heard from 6 Andy> people who are against this, but don't want their identities Andy> known, so the proposal is rejected." Not acceptable. I think that would be unacceptable. I think that a WG chair going to people who expressed private concerns and saying something like "Hey, you need to express your concerns in public. They are shared; if all of the people who have these concerns bring them forward then we would have enough interest in dealing with this issue. You have a week," is entirely fine. I also think it is fine for a WG chair to look at private technical concerns, realize they are correct and raise them to the WG. "I received a private concern; that mail pointed out that the following trivial attack will break the security of this protocol. We are not moving forward until someone fixes this problem or someone explains why I'm misunderstanding the situation." It's probably even fine to say "I received a lot of private concerns. Are the people willing to make public comments firmly behind their support?" --Sam ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Sam Hartman wrote: "Andy" == Andy Bierman <[EMAIL PROTECTED]> writes: Andy> This is not an alternative. If you are not willing to make Andy> your technical objections to a technical specification Andy> publicly, then they cannot be part of the IETF Andy> decision-making process. At one level I agree here. Andy> What's to prevent a WG Chair from "padding" the anonymous Andy> "votes"? If 5 people in public (WG meeting or mailing list) Andy> are for some proposal, and the Chair says, "I heard from 6 Andy> people who are against this, but don't want their identities Andy> known, so the proposal is rejected." Not acceptable. I think that would be unacceptable. I think that a WG chair going to people who expressed private concerns and saying something like "Hey, you need to express your concerns in public. They are shared; if all of the people who have these concerns bring them forward then we would have enough interest in dealing with this issue. You have a week," is entirely fine. I also think it is fine for a WG chair to look at private technical concerns, realize they are correct and raise them to the WG. "I received a private concern; that mail pointed out that the following trivial attack will break the security of this protocol. We are not moving forward until someone fixes this problem or someone explains why I'm misunderstanding the situation." I don't understand why such a comment needs to be private. Once the issue comes to light in the WG, it is no longer going to be private. You are assuming the Chair can and should be a proxy for a WG member who wishes to remain anonymous. I disagree. It's probably even fine to say "I received a lot of private concerns. Are the people willing to make public comments firmly behind their support?" I am specifically referring to technical comments. I realize that WG members may have non-technical concerns which are appropriate to convey to the Chair privately. I think the IETF consensus process is severely flawed. Many times I have encountered deadlocks because 3 people are strongly for something, 3 people are strongly against it, and 40 people couldn't care less which way the decision goes. Determining consensus based on hearsay and humming makes matters even worse. --Sam Andy ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Andy Bierman wrote: Michael Thomas wrote: I think the inability of the IETF to make decisions in an open, deterministic, and verifiable manner is a major flaw. It promotes indecision and inaction. Is there any human decision making process that has all of these characteristics? Or that even believes that those are axiomatic? Dishonesty by the management is a problem regardless of what system we have. Most wg's these days have two chairs, so collusion would need to be at least that deep, and probably require an AD to be on board too. If that really were the case, I doubt any system is likely to perform very well. Only transparency can prevent corruption. Preventing corruption is not the end product of the IETF. Producing good/useful protocol specs is the end product of the IETF. Thus "corruption" is just one consideration. Life is messy that way. But this cultural thing does bug me. It seems unsatisfying to me that our pat answer to cultural differences is "become more western". The language issue is already asking quite a lot of the rest of the world. I don't see the cultural bias here. Which culture are you from? Mike ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Michael Thomas wrote: Andy Bierman wrote: Michael Thomas wrote: I think the inability of the IETF to make decisions in an open, deterministic, and verifiable manner is a major flaw. It promotes indecision and inaction. Is there any human decision making process that has all of these characteristics? Or that even believes that those are axiomatic? no, but there is some evidence that the IETF does not regularly make timely and effective decisions, relative to the expectations of the participants. IMO, this is due to a flawed process which only works well in "landslide" decisions. The track record for close or difficult decisions is not very good. Dishonesty by the management is a problem regardless of what system we have. Most wg's these days have two chairs, so collusion would need to be at least that deep, and probably require an AD to be on board too. If that really were the case, I doubt any system is likely to perform very well. Only transparency can prevent corruption. Preventing corruption is not the end product of the IETF. Producing good/useful protocol specs is the end product of the IETF. Thus "corruption" is just one consideration. Life is messy that way. The end product is a document. The document is a result of decisions made (or punted). The text in the document is fair game for any kind of comment, as long as the comment is made openly for everyone in the WG to consider. If there was an anonymous mailing service, such that people could comment on WG issues without revealing their identity, then how do we know everybody using the service will only make their comment once? What is to prevent them from making the same comment N times, hoping to deceive the WG into thinking the comments represent the views of N people? But this cultural thing does bug me. It seems unsatisfying to me that our pat answer to cultural differences is "become more western". The language issue is already asking quite a lot of the rest of the world. I don't see the cultural bias here. Which culture are you from? Kalifornia -- as if that matters. The IETF is not an academic exercise. I have found that people from the academic culture are appalled at how their documents are treated by others in the IETF. In the competitive culture of the IETF, documents are often attacked (or ignored) in ways that are not common in the academic world. This is unfortunate, but not enough of a reason to allow the IETF decision making process to be even more murky and secretive. Mike Andy ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Maybe I have misread the exchange. But I do expect chairs to receive private comments about the state of things. And to try to respond helpful to those comments when they can. And I expect them to make use of that exchange to help the public conversation. To use a current example, the chair of one working group just said (paraphrasing) I have received several comments privately that the description of X in the document doesn't match reality. However, only person A has spoken up on the list. If this is a real issue, please speak up on the list NOW. Being realistic, there are people who will be reluctant to speak up on the list. particularly when some of our best and most active participants have been known to land on folks who disagree with them like a ton of bricks. The chair clearly can not let the private comments change the public consensus. But he can use it to be able to tell people that several folks agree, and manage the process. There are also other reasons for sending comments to the chair. For example, I have sent comments saying roughly ~if I raise X it will cause a lot of email exchange without resolving anything. As long as you, chair, think we have discussed it enough I will refrain from raising it.~ And many other reasons come up. Yours, Joel At 07:14 PM 5/31/2007, Andy Bierman wrote: I don't understand why such a comment needs to be private. Once the issue comes to light in the WG, it is no longer going to be private. You are assuming the Chair can and should be a proxy for a WG member who wishes to remain anonymous. I disagree. ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
The current process doesn't work very well when voting is required, after hum-style consensus has been inconclusive. I think a fair vote requires - a clear definition of who can vote - a vote that is announced well in advance to all parties, not just a select few - some process that avoids favoring one or the other group (e.g., by holding a vote during a meeting, while allowing remote participation, but holding the vote at such a time that only certain regions can reasonably participate) These difficulties are particularly pronounced when a vote is held at a meeting, but the issues of franchise apply in general. [Without getting into the details, let's just say that I had the non- pleasure of participating in a WG vote that had issues with all of the above.] Henning On May 31, 2007, at 3:02 PM, Melinda Shore wrote: On 5/31/07 2:49 PM, "John C Klensin" <[EMAIL PROTECTED]> wrote: I think this is more evidence that we need more flexibility and good sense, not more rigid rules. Well, what's under description really isn't consensus decision-making processes - what's being argued is a sort of voting. Rather than getting stuck on definitions, though, I think it's probably worthwhile to frame the discussion in terms of intentions - what the original intent was, what the current intent is, whether or not there's a mismatch, and how to get the process to look like what's intended, if it currently doesn't (and wow, do I think it does not). Arguing over percentages doesn't, I think, go very far towards figuring those things out. I think I understand what used to be intended but I'm not sure I can articulate it, but I do think that the problems around decision-making come down to a few things: 1) not-that-great decision facilitation skills on the part of some chairs and leaders; 2) too many participants who'd be happy shutting down the whole process rather than accomodating a decision they don't agree with 3) the organization is just too large for a touchy- feely decision-making style. But it seems to me that if you're going to go with some form of voting, and that's what's happening in practice, it would be better to design a fair voting process. If you're going try to do consensus better you have to figure out how to deal with the kvetches in a way that doesn't shut out dissent while still allowing decision-making to move forward. Either one is hard to do but I'm not sure that doing nothing is an option. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: consensus and anonymity
Can anyone point to me where it is written that voting at a meeting is the decision making process when rough consensus (hum or whatever) has been inconclusive? From what I know and have been told, people on the list should have been given the chance to opine before declaring any results (so the issue of timezones and advanced notice should not have any impact). Could the IESG/IAB confirm what is the right procedure? thanks, Lakshminath On 5/31/2007 7:09 PM, Henning Schulzrinne wrote: The current process doesn't work very well when voting is required, after hum-style consensus has been inconclusive. I think a fair vote requires - a clear definition of who can vote - a vote that is announced well in advance to all parties, not just a select few - some process that avoids favoring one or the other group (e.g., by holding a vote during a meeting, while allowing remote participation, but holding the vote at such a time that only certain regions can reasonably participate) These difficulties are particularly pronounced when a vote is held at a meeting, but the issues of franchise apply in general. [Without getting into the details, let's just say that I had the non-pleasure of participating in a WG vote that had issues with all of the above.] Henning On May 31, 2007, at 3:02 PM, Melinda Shore wrote: On 5/31/07 2:49 PM, "John C Klensin" <[EMAIL PROTECTED]> wrote: I think this is more evidence that we need more flexibility and good sense, not more rigid rules. Well, what's under description really isn't consensus decision-making processes - what's being argued is a sort of voting. Rather than getting stuck on definitions, though, I think it's probably worthwhile to frame the discussion in terms of intentions - what the original intent was, what the current intent is, whether or not there's a mismatch, and how to get the process to look like what's intended, if it currently doesn't (and wow, do I think it does not). Arguing over percentages doesn't, I think, go very far towards figuring those things out. I think I understand what used to be intended but I'm not sure I can articulate it, but I do think that the problems around decision-making come down to a few things: 1) not-that-great decision facilitation skills on the part of some chairs and leaders; 2) too many participants who'd be happy shutting down the whole process rather than accomodating a decision they don't agree with 3) the organization is just too large for a touchy- feely decision-making style. But it seems to me that if you're going to go with some form of voting, and that's what's happening in practice, it would be better to design a fair voting process. If you're going try to do consensus better you have to figure out how to deal with the kvetches in a way that doesn't shut out dissent while still allowing decision-making to move forward. Either one is hard to do but I'm not sure that doing nothing is an option. Melinda ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
Re: On the IETF Consensus process
Good points. I don't have a miracle cure, but I am willing to take the risk to point out things that are not going well and what might need to be done to make things more transparent, increase accountability and implement some checks and balances. I identified two things: 1. The IAB to make BoF reports public; 2. Implement more checks and balances. I will start threads on the two topics soon. thanks, Lakshminath On 5/27/2007 8:18 AM, Eastlake III Donald-LDE008 wrote: Lakshminath, The WG Chairs and Ads are all human beings, are all skewed to some extent, and all have some philosophy whether strong or weak. I don't see how this can be avoided. The effect of this may not be what you expect. For example, one of these, realizing they have a bias towards X, may try to compensate and adjust their decisions so as to end up having an anti-X effect. Hopefully the element of randomness in the nomcom process decreases the correlation in the bias of successive IESG members. Assume you could come up with some metric and say that an official's decisions were 5 percent biased. Perhaps, by devoting an ever increasing amount of time and effort to controlling this bias you could reduce it. But every ounce of effort you apply in that direction is unavailable to contribute to technical efforts. And it is not clear how you measure bias in an un-biased way since everyone will have a different opinion of what biases there are and how big they are. In fact, as you continue to increase the time and effort being put into "controlling bias", you end up with things being controlled by those who have time and effort available to try to control the organization rather than make technical contributions. If you can find a miracle cure that improves IETF decision making while decreasing the informational and decision making load of the members of the IETF community, that would be great, but I'm dubious. Thanks, Donald -Original Message- From: Jeffrey Hutzelman [mailto:[EMAIL PROTECTED] Sent: Thursday, May 24, 2007 2:11 PM To: Lakshminath Dondeti Cc: Jeffrey Hutzelman; IETF Discussion Subject: Re: On the IETF Consensus process On Wednesday, May 23, 2007 06:56:10 PM -0700 Lakshminath Dondeti <[EMAIL PROTECTED]> wrote: Hi Jeff, ... Consider what happens if a WG chair or an AD's decisions are skewed, either intentionally or because they are naturally biased towards a particular philosophy? Often people tend to try and live with it or adjust to it. There is not really a viable avenue to provide feedback about the AD. Appealing (happens rarely) or recalling (never happened?) are drastic measures. Yes, they are, and no, the recall procedure has never been used. I'm sort of torn on this - most every decision is appealable, and if an AD is making bad decisions and won't listen to reason, they should be appealed. However, if everyone appealed every decision they didn't like, we'd never get anything done. ... -- Jeff ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf ___ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
