| RFC 9400 | Organization of Online Meetings | June 2023 |
| Kühlewind & Duke | Informational | [Page] |
This document provides guidelines for the planning and organization of fullyonline meetings, regarding the number, length, and composition of sessions onthe meeting agenda. These guidelines are based on the experience gained by holding online meetings during the COVID-19 pandemic in 2020 and 2021.¶
This document is not an Internet Standards Track specification; it is published for informational purposes.¶
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are candidates for any level of Internet Standard; see Section 2 of RFC 7841.¶
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained athttps://www.rfc-editor.org/info/rfc9400.¶
Copyright (c) 2023 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
In 2020, the COVID-19 pandemic forced the IETF to convert all its plenary meetingsto online-only events. This document records the experience gained by holdingplenary meetings fully online and proposes guidelines based on this experience.In general, participant surveys indicated satisfaction with the organization ofthese meetings.¶
Although these guidelines reflect lessons learned in 2020 and 2021, the IETF isencouraged to continue to experiment with the format and agenda of fully onlinemeetings, using this document as a baseline.¶
Hybrid meetings (meaning meetings that have large remote participation but alsoonsite participation) are out of scope. However, some of the experience gainedfrom fully online meetings might also provide input for decisions regarding theorganization of hybrid meetings.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14[RFC2119][RFC8174] when, and only when, they appear in all capitals, as shown here.¶
This document uses the term "plenary meeting" for the whole IETF meeting thatcovers the IETF meeting week; this term is used to distinguish the plenary meetingfrom other IETF meetings like "interim meetings".The term "administrative plenary" is used for the respective sessionduring the IETF meeting week that is usually hosted on Wednesday.¶
When the World Health Organization (WHO) declared a worldwide pandemicin March 2020, the IETF canceled itsplenary meeting and organized an online replacement in less than 2 weeks. Forthis first online-only meeting, the agenda was reduced to a set of sessions thatbenefited most from cross-area participation, like BoFs, first-time meetings ofnew working groups, and dispatch sessions. It also included the administrativeplenary to preserve the official handover procedures that occur at March IETFmeetings, as described in[RFC8713].¶
With a reduced agenda, the meeting format was two sessions (about 4 hours) per daywith a maximum of two parallel tracks. Other working group meetings werescheduled as interims over the following 6 weeks. The IESG published apurely advisory recommended schedule[INTERIM-SCHEDULE] to reduce conflictsamong those interims.¶
While satisfaction was high right after the meeting[IETF107-FEEDBACK], some participantslater indicated in mailing list discussions that the period of intensive interims had agreater impact on their calendar than a single plenary meeting week, and in somemeetings participation was reduced. Those interims tended to occur at times convenient for the bulkof participants, which was convenient for most but could exclude those in lesscommon time zones.¶
For the remainder of 2020 and 2021, the online schedule was switched back to besimilar to an in-person meeting (1- to 2-hour slots and eight or nine parallel tracks).However, each day was limited to 5-6 hours in recognition that remoteparticipation is more tiring.¶
All fully online meetings followed the time zone of the planned in-personmeeting location. As a 6-hour agenda has some flexibility regarding the starttime while still fitting within a previously used 8-hour in-person agenda, the starttime was approximately noon, with adjustments of an hour or so to mitigate theimpact of early morning hours in time zones with many participants. As selectionof in-person meeting sites was consistent with the 1-1-1 guideline as documentedin[RFC8719], this approach was intended to share the burden across all commongeographies roughly equally.¶
The following algorithm was not used in 2020 or 2021, but it enables mostparticipants to avoid late-night sessions in two out of every three fully online IETFplenary meetings. Basically, every fully online meeting is for two regions of thethree regions described in[RFC8719], with one being roughly after sunriseand the other around sundown. This has the trade-off that the third region is inthe middle of night.¶
The times are also seasonally adjusted to leverage differentials in DaylightSaving Time. These time slots are as follows, in UTC, based on the DaylightSaving Practices at the time of publication:¶
| Name | Times (Northern Summer) | Times (Northern Winter) |
|---|---|---|
| North America Night | 0500-1100 UTC | 0600-1200 UTC |
| Asia Night | 1300-1900 UTC | 1400-2000 UTC |
| Europe Night | 2200-0400 UTC | 2200-0400 UTC |
Note that the "Europe Night" slot covers the "early morning" slot for Asia wheremost countries do not have Daylight Saving Time.¶
If Daylight Saving Practices change -- this change is under consideration in multiple countriesat the time of publication -- this table may need adjustment.¶
The intent of rotating between these three slots is to scatter meetingsthroughout the course of the global day, to maximize the ease of participantsso that no attendee has to beconsistently inconvenienced, regardless of their location and what time of day isoptimal for their schedule. However, as participation is distributed globally,it needs to be acknowledged that restricting the scheme to three regionsobserves the intent of[RFC8719] but does not achieve thegoal of two non-late-night sessions for all participants equally.¶
The IETFSHOULD select a start time from these three choices based on the prior three meetings. The following table covers all permutations of previousmeetings held in person in Region A, B, or C or remotely in the nights ofone of those regions.¶
| Three Meetings Ago | Two Meetings Ago | Last Meeting | Online Selection |
|---|---|---|---|
| Any | Any | In-Person A | A Night |
| Any | Online A Night | Online B Night | C Night |
| Online A Night | In-Person B | Online B Night | C Night |
| In-Person A | In-Person B | Online B Night | A Night |
| In-Person A | In-Person A | Online A Night | See below |
| Online A Night | Online B Night | Online C Night | A Night |
This table follows two basic guidelines:¶
The final case occurs in the rare event that back-to-back in-person plenary meetingsoccur in the same region. In this case, find the most recent meeting that wasin neither 'A' (if in person) nor 'A Night' (if fully online). If this meetingwas in person in region 'B', then the next meeting should be in 'B Night'. If itwas remote in 'B Night', the next meeting should be in 'C Night'.¶
By 2021, fully online meetings were consistently held over 5 days with roughly 6-hourmeeting days. The day with the administrative plenary, which concludes withmultiple open mic sessions, sometimes exceeded this limit.¶
Six hours of online meetings, with two 30-minute breaks, was a compromisebetween the physical limits of attending an online meeting in an inconvenienttime zone and the demand for many sessions with a manageable number ofconflicts. The IETF 109 feedback[IETF109-SURVEY] indicated broad satisfactionwith a 5-day meeting but only medium satisfaction with the overall length ofeach day.¶
The IETF did not seriously consider extending sessions into the weekend beforeor after the main meeting week, although at IETF 108 and subsequent meetings the Hackathon occupied the entire week before (see[RFC9311]).¶
For fully online meetings, there are typically fewer sessions per day than forin-person meetings, to keep the overall meeting day to roughly 6 hours.With fewer sessions, chairs were offered only two options for session length(instead of three).¶
IETF 108, based on an indicated preference of the community, scheduled 50- and100-minute slots, with 10-minute breaks, in order to keep the overall day lengthat 5 hours. This resulted in many sessions going over time, which indicated that10 minutes for breaks is not practical.¶
The survey after IETF 109[IETF109-SURVEY] showed high satisfaction with 60/120-minutesession lengths and 30-minute breaks, and a significant improvement insatisfaction over IETF 108.¶
The longer breaks, while extending the day, provided adequate time for meals,exercise, and "hallway" conversations using online tools.¶
In-person meetings are limited in the number of parallel tracks by the number ofmeeting rooms, but online meetings are not. However, more parallel trackswould increase the number of possible agenda conflicts.¶
If the total number of requested sessions exceeds the capacity of the usual eightparallel tracks, it is possible for a fully online meeting to simply use moretracks. If the number and length of meeting days are seen as fixed, this decisionis implicitly made by the working group chairs requesting a certain number ofsessions and length.¶
IETF 111 used nine parallel tracks for some of the sessions and experiencedslightly more conflicts in the agenda-scheduling process, though there was nostatistically significant increase in dissatisfaction about conflicts in thesurvey[IETF111-SURVEY].¶
The IESG encouraged working group chairs to limit their session requests and useinterim meetings aggressively for focused work.¶
The IETF 108 meeting survey[IETF108-SURVEY] asked about the structure of thatmeeting (full meeting) compared to that of IETF 107, which hosted only a limitedset of sessions followed by interims in the weeks after. The structure of IETF108 was preferred by 82%. Respondents valued cross-participation and anintensive meeting week for maintaining project momentum.¶
Furthermore, a well-defined meeting time, rather than spreading many interimsover the whole year, can make deconflicting with other non-IETF meetings easier.¶
However, interim meetings can also help to reduce scheduling conflicts during anIETF week and allow for a more optimal time slot for the key participants. Whileinterim meetings are less likely to attract people with casual interest, theyprovide a good opportunity for the most active participants of a group to havedetailed technical discussions and solve recorded issues efficiently.¶
This document recommends further experiments with reducing conflicts byleveraging the increased flexibility of the online format.¶
An in-person meeting must fit all sessions into an acceptable length forinternational travel (usually roughly a week), but online meetings do not havethat constraint.¶
Therefore, it would be possible to keep most regular working group sessionswithin the usual 5 main meeting days but have some of the more conflictedsessions in other dedicated time slots. As the Hackathon for fully onlinemeetings is usually held in the week before the online plenary meeting[RFC9311], that week is already a highly active week for many IETFparticipants and might provide an opportunity to schedule a few selectedsessions.¶
This might work especially well for sessions that are of high interest to alarge part of the community, such as BoFs and dispatch meetings, and therefore hardto schedule during the main IETF week.¶
At IETF 112, the IESG ran an experiment where the administrative plenary wasscheduled on the Wednesday before the official session week. The experimentreport[IETF112-EXPERIMENT] found that it led to a reduction in schedulingconflicts but also a slight drop in attendance of the administrative plenary, partlydue to insufficient awareness.¶
Participation in the fully online meetings in 2021 was high and had a stableper-country distribution, even though time zones were rotated. This indicatesthat online meetings support a more consistent geographic distribution ofparticipants than in-person meetings, where participation often fluctuates basedon the location.¶
However, online meetings do not provide an equivalent opportunity to socialize.Despite significant investment in tools to foster hallway conversations, manydid not use those tools, whether due to ignorance of them, dislike of the tools,or a preference for other activities at home (including sleep and food)over hallway interactions.¶
There was a decrease in submissions of new (-00) Internet-Drafts during 2020 and2021, although the overall number of draft submissions remained stable;this decrease in new submissions might have resulted from the loss of these interactions. Informal conversations mightbe important to inspire new work.¶
This document recommends further experiments with the meeting structure. Often,only practical experience can answer open questions. A given meetingSHOULD onlyexperiment with one major change at a time in order to be able to assess the outcome correctly.Furthermore, the IESGSHOULD announce any such experiment well in advance, so peoplecan adjust to changes and potentially provide feedback.¶
This document has no IANA actions.¶
This document has no security considerations.¶
Thanks toBrian Carpenter,Lars Eggert,Toerless Eckert,Charles Eckel,Jason Livingood,Sanjeev Gupta,Dale Worley, andMark Nottingham for theirreviews, and thanks to the many other people who provided input and suggestions on the time zone discussion!¶