I was about to respond to Brian's note, but then found Keith's, which covers some of what I was going to say although I disagree with him on other parts of the issue. Inline. --On Friday, May 10, 2019 19:45 -0400 Keith Moore <moore@xxxxxxxxxxxxxxxxxxxx> wrote: > On 5/10/19 4:33 PM, Brian E Carpenter wrote: > >>> And this has been a problem since the early 1990s when the >>> US government stopped subsidizing the meetings (and perhaps >>> also the secretariat?). But I wish we'd try harder to find >>> that sustainability model rather than constantly punting the >>> problem, because the Internet has been suffering for all >>> that time from a lack of diverse participation in IETF. >> I don't see how the IETF is supposed to fix the fact that >> independent open source developers are, um, independent. > > This reply seems to presume that "independent" developers > should be considered by IETF to be the exceptional case, and > only "dependent" developers (presumably those funded by huge > corporations) have a right to sit at the big table. I > think it should be the other way around - IETF should be > optimized to facilitate contributions from independent > parties, and those with sponsorship are welcome to sit at the > same table as everyone else. I think optimizing for either is fairly hopeless. "Independent" presumably includes open source developers, contributors who support themselves as consultants or the like with multiple clients with connections to IETF work, contributors who similarly support themselves but whose clients are mostly not connected to the IETF, academics with research funding that can cover travel and/or at least some meeting costs, academics without such funding, and probably several other groups. We (or at least ISOC) have also made distinctions between people from developing countries and ones with more developed ones (and I recall as discussion or two about the meaning of "from" in that context). At the same time, we continue to claim that all IETF participation --at least as far as taking of positions and determination of consensus are concerned-- is by individuals. That is a myth, I think increasingly so, and I fear that it will sooner or later get us in trouble but moving away from it would be a big step. > There was a time when IETF was more like this, even after we > had to pay our own meeting costs. We got sucked into the > mode of holding meetings at expensive hotels, especially after > our attendance figures pushed into the multiple thousands and > they were the only venues available other than conference > centers (which were even more expensive). It got even > harder after IETF started meeting in more diverse locations > (which it should have done, and which I supported). But I > also saw various efforts to raise the bar on participation and > discourage the independents. I'm fairly cynical, but I've nonetheless seen little evidence that any efforts to discourage independents were deliberate. What I have seen would be consistent with people who have had strong organizational financial (and other) support for meeting attendance and participation losing track of the consequences of decisions on those who lack such support and for whom even relatively small incremental costs may be significant. As examples, I note that the pushback from "independents" against expanding meetings to include Fridays were largely dismissed. Various efforts to limit the number of WGs or raise the threshold for starting them or continuing ones that are not performing (which could have helped hold down costs as well as the size of the IESG) having gotten traction either, I tend to assume for the same reason. We've also very significantly increased the number of interim meetings in recent years and there has been, AFAICT, no effort to question the appropriateness or necessity of those and balance their importance against the risks of making is much more difficult for some members of the community to participate. >> It seems to me that the current focus on improving remote >> attendance facilities is really the best we can do, but >> again: if remote attendance really becomes as good as on-site >> attendance, the number of funded atttendees will rapidly >> decline. >> I think that if there was a viable answer to this problem, >> we'd already have found it. > > I wonder whether if the organization had looked for an answer, > we'd have already found it. I would have said "looked in the right places". For example, if remote participation works (as I have found it does if one understands and accepts the limitations), the number of Wg-specific interims is on the increase, and we are (I believe) seeing less actual cross-area in-depth technical review than we did a decade ago, perhaps those factors are enough of our present-day reality that we could consider dropping from three f2f plenary-type meetings a year to two, a move that, all other things being equal, would result in an immediate 33% drop in costs to participants and, unlike many other proposals, would presumably produce a significant drop in Secretariat and other overhead costs. There have been a few suggestions to consider that too and they have gone nowhere (and that doesn't mean "the community discussed them and reached rough consensus"). > But conditions have changed and keep changing. Remote > participation is more feasible today than in the past. And > our attendance numbers are down from the late 1990s period > which should give us some additional flexibility about venues. > I wonder if it's possible to rely less on face-to-face > meetings, while still having them (because personal contact is > still very valuable to enabling people to collaborate at a > distance), and make the face-to-face meetings less expensive > (maybe by making them shorter and more focused on facilitating > the personal interaction that really helps people work well > remotely.) Yes, and see above. But despite my preference for four-day meetings over five (if only to avoid killing two consecutive weekends for most participants), I'm guessing that most of the costs of holding a meeting (both overhead and to participants) are associated with holding the meeting, i.e., that while chopping off a day or two would save money, it would save a lot less than the proportion of that time to current total meeting time. > A tremendous amount of participants' time is wasted at our > current meetings by making Powerpoint (and similar) > presentations the normal mode of working group sessions. > Before we used Powerpoint, our in-person discussions were much > more productive. And that could help us make the > meetings shorter and more useful at the same time. I can't add much to what others have said about this. However, although I have many days when I believe that overhead projectors and whiteboards were far superior for interactive presentations and discussions than video presentations of slides (PowerPoint or otherwise), I don't see any way to get back to those times. I also see efficient real-time drawing or editing by remote participants as being unlikely, so there may also be a tradeoff involved. > Perhaps we can't completely level the playing field but > there's plenty of room to do better. Yes. I think the prerequisite to significant improvement is a change in attitude and increased awareness of the degree to which costs, even at present levels, are increasing the homogeneity of the IETF and biasing against or excluding perspectives that are badly needed. The oft-repeated saying is "tell it to the Nomcom" but the same kinds of costs, as well as our current eligibility criteria, tend to create Nomcoms that are highly biased toward those with strong organizational (usually corporate) support, many of whom are not sensitive enough how the associated assumptions change the IETF's effective composition and that of its leadership bodies. Similarly, if a particular body tells the Nomcom that someone cannot be an effective member of that body unless they can and will attend three full IETF meetings a year (probably for seven rather than five days), another several days in one or two "retreats", and perhaps some f2f interim meetings -- I hope I'm exaggerating, but fear that I'm not-- perhaps the community should be pushing back and asking hard questions about the value of those requirements versus the effect of excluding almost anyone who does not have full-time corporate support and essentially unlimited travel budget for those roles. These are obviously much broader questions than ones about support for a few extra people to get to a meeting, but I don't see how to answer the narrower question fairly without addressing at least some of the broader ones. john
