| Network Working Group | D. Crocker |
| Internet-Draft | Brandenburg InternetWorking |
| Intended status: Informational | N. Clark |
| Expires: January 21, 2016 | Pavonis Consulting |
| July 20, 2015 |
An IETF with Much Diversity and Professional Conduct
draft-crocker-diversity-conduct-06
The process of producing today's Internet technologies, through a culture of open participation and diverse collaboration has proved strikingly efficient and effective, and it is distinctive among standards organizations. For its early years, participation in the IETF and its antecedent was almost entirely composed of a small group of well-funded, American, white, male technicians, demonstrating a distinctive and challenging group dynamic, both in management and in personal interactions. In the case of the IETF, interaction style can often contain singularly aggressive behavior, often including singularly hostile tone and content. Groups with greater diversity make better decisions. Obtaining meaningful diversity requires more than generic good will and statements of principle. Many different behaviors can serve to reduce participant diversity or participation diversity. This document discusses IETF participation, in terms of the nature of diversity and practical issues that can increase or decrease it. The document represents the authors' assessments and recommendations, following general discussions of the issues in the IETF.
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
This Internet-Draft will expire on January 21, 2016.
Copyright (c) 2015 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 (http://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 Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
This document discusses IETF participation, in terms of the nature of diversity and practical issues that can increase or decrease it. The topic has received recent discussion in the IETF, and the document represents the authors' assessments and recommendations about it, in the belief that it is constructive for the IETF and that it is consonant with at least some of the IETF community's participants.
The Internet Engineering Task Force [IETF] grew out of a research effort that was started in the late 1960s, with central funding by the US Department of Defense Advanced Research Projects Agency (ARPA, later DARPA) employing a collection of research sites around the United States, and including some participation by groups of the US Military. The community was originally restricted to participation by members of the funded research groups. In the 1980s, participation expanded to include projects funded by other agencies, most notably the US National Science Foundation for its NSFNet effort. At around the time the IETF was created in its current form, in the late 1980s, participation in the group became fully open, permitting attendance by anyone, independent of funding, affiliation, country of origin, or the like.
Beyond the obvious effects of the resulting technology that we now enjoy, the process of producing today's Internet technologies through a culture of open participation and diverse collaboration has proved strikingly efficient and effective, and it is distinctive among standards organizations. This culture has been sustained across many changes in participant origins, organizational structures, economic cycles, and formal processes. However maintenance of the IETF's effectiveness requires constant vigilance. As new participants join the IETF mix, it is increasingly easy for the IETF's operation to gradually invoke models from other environments, which are more established and more familiar, but often are less effective.
Historically participation in the IETF and its antecedent was almost entirely composed of a small group of well-funded, American, white, male technicians. No matter the intentions of the participants, such a narrow demographic demonstrated a distinctive group dynamic, both in management and in personal interactions, which persists into the current IETF. Aggressive and even hostile discussion behavior is quite common. In terms of management the IETF can be significantly in-bred, favoring selection of those who are already well-known. Of course, the pool of candidates from which selections are made suffer classic limitations of diversity found in many engineering environments. Still there is evidence and perception of selection bias, beyond this.
In the case of the IETF, the style of interaction can often demonstrate singularly aggressive behavior, including singularly hostile tone and content. In most professional venues, such behavior is deemed highly unprofessional, or worse. Within the IETF, such behavior has had long-standing tolerance. Criticizing someone's hostility is dismissed by saying that's just the way they are, or that someone else provoked it, or that the person is generally well-intentioned. Further anyone expressing concern about the behavior is typically admonished to be less sensitive; that is, a recipient of an attack who then complains is often criticized or dismissed.
As the IETF opened its doors to participation by anyone, its demographics have predictably moved towards much greater variety. However the group culture has not adapted to accommodate these changes. The aggressive debating style and the tolerance for personal attacks can be extremely off-putting for participants from more polite cultures. And the management selection processes can tend to exclude some constituencies inappropriately.
Recently, members of an informal IETF women's interest group, called "systers", organized a quiet experiment, putting forward a large number of women candidates for management positions, through the IETF's "Nomcom" process. Nomcom is itself a potentially diverse group of IETF participants, chosen at random from a pool of recent meeting attendees who offer their services. Hence its problematic choices -- or rather, omissions -- could be seen as reflecting IETF culture generally.
Over the years some women have been chosen for IETF positions as authors, working group chairs, area directors, Internet Architecture Board [IAB] members and IETF Administrative Oversight Committee[IAOC] members. However the results of the systers experiment were not encouraging. In spite of their recruiting a disproportionately high number of female candidates, not a single one was selected. Although any one candidate might be rejected for entirely legitimate reasons, a pattern of rejection this consistent suggested an organizational bias. The results were presented at an IETF plenary and it engendered significant IETF soul-searching, as well as creation of a group to consider diversity issues for the IETF.[Div-DT][Div-Discuss]
Other activities around that same time also engendered IETF consideration of unacceptable behaviors, generally classed as harassment. This resulted in the IESG's issuing a formal IETF anti-harassment policy.[Anti-Harass]
Changing an organization's culture is difficult and requires not only commitment to the underlying principals, but also vigilant and sustained effort. The IESG has taken essential first steps. What is needed is going beyond the position papers and expression of ideals, into continuing education of the entire community, and immediate and substantive response to unacceptable behaviors.
Diversity concerns the variability of a group's composition. It can reasonably touch every conceivable participant attribute. It includes task-related attributes, such as knowledge and experience, as well as the usual range of "identified class" attributes, including race, creed, color, religion, gender and sexual orientation, but also extends along with all manner of beliefs, behaviors, experiences, preferences and economic status.
The factors affecting the quality of group decision-making are complex and subtle, and are not subject to precise specification. Nevertheless in broad terms, groups with greater diversity make better decisions.[Kellogg] They perform better at diverse tasks both in terms of quantity and quality and a great deal of research has found that heterogeneity often acts as a conduit for ideas and innovation.[WiseCrowd],[Horowitz],[Stahl],[Joshi] The implicit assumptions of one participant might not be considerations for another, and might even be unknown by still others. And different participants can bring different bases of knowledge and different styles of analysis. People with the same background and experience will all too readily bring the same ideas forward and subject them to the same analysis, thus diminishing the likelihood for new ideas and methods to emerge, or underlying problems to be noted.
However a desire to diligently attend to group diversity often leads to mechanical, statistical efforts to ensure representation by every identified constituency. For smaller populations, like the IETF and especially for its small management teams, this approach is counter-productive. First, it is not possible to identify every single constituency that might be relevant. Second, the group size does not permit representation by every group. Consequently, in practical terms, legitimate representation of diversity only requires meaningful variety, not slavish bookkeeping. In addition, without care it can lead to the negative effects of diversity where decision making is slowed, interaction decreased and conflict increased.[Horowitz]
Pragmatically, then, concern for diversity merely requires serious attention to satisfying two requirements:
In other words, look for real variety in group composition and real variety in participant discussion. This will identify a greater variety of possible and practical solutions.
Obtaining meaningful diversity requires more than generic good will and statements of principle. The challenges, here, are to actively:
It also requires education about the practicalities of diversity in an open engineering environment; and it requires organizational processes that regularly consider what effect each decision might have on diversity.
Examples abound:
Many different behaviors can serve to reduce participant diversity or participation diversity. One class of efforts is based on overt actions to marginalize certain participants, by intimidating them into silence or departure. Intimidation efforts divide into two styles warranting distinction. One is harassment, which pertains to biased treatment of demographic classes. A number of identified classes are usually protected by law and community understanding that such biased behavior can not be tolerated has progressively improved.
Other intimidation efforts are tailored to targeted individuals and are generally labeled bullying.[Har-Bul],[Workplace],[Signs], [Escalated], [Prevention] The nature and extent of bullying in the workplace is widely underestimated, misunderstood and mishandled. It is:
Whether directed at classes or individuals, intimidation methods used can: [Dealing]
If tolerated by others, and especially by those managing the group, these methods create a hostile work environment.
The IETF's Anti-Harassment Policy [Anti-Harass] uses a single term to cover the classic harassment of identified constituencies, as well as the targeted behavior of bullying. The policy's text is therefore comprehensive, defining unacceptable behavior as "unwelcome hostile or intimidating behavior." Further it declares: "Harassment of this sort will not be tolerated in the IETF." An avenue for seeking remedy when harassment occurs is specified as a designated Ombudperson.
Unified handling of bullying and harassment is exemplified in the policies of many different organizations, notably including those with widely varying membership, even to the point of open, international participation, similar to that of the IETF. Examples include:
In fact there is a view that harassment is merely a form of bullying, given the same goal of undermining participation by the target:
The IETF has a long history of tolerating aggressive and even hostile behavior by participants. So this policy signals a formal and welcome change. The obvious challenge is to make the change real, moving the IETF from a culture that tolerates -- or even encourages -- inter-personal misbehaviors to one that provides a safe, professional, and productive haven for its increasingly-diverse community.
Here again, examples abound, to the present:
The goal of open, diverse participation requires explicit and on-going organizational effort, concerning group access, engagement and facilitation.
Aiding participants with access to IETF materials and discussions means that it is easy for them to:
After materials and discussions are located, the primary means of making it easy to access the substance of the work is for statements to be made in language that is clear and explanatory. Writers and speakers need to carefully consider the likely audience and package statements accordingly. This often means taking a more tutorial approach than one might naturally choose. In speech, it means speaking more deliberately, a bit more clearly and a bit more slowly than needed with close collaborators. When language is cryptic or filled with linguistic idiosyncrasies and when speech is too fast, it is dramatically less accessible to a diverse audience.
Once content is accessible, the challenge is to garner diverse contribution for further development. Engagement means that it is easy for constructive participants to be heard and taken seriously through constructive interaction.
Within the IETF, the most common challenge is choosing how to respond to comments. The essence of the IETF is making proposals and offering comments on proposals; disagreement is common and often healthy... depending upon the manner in which disagreement is pursued.
In order to obtain the best technology, the best ideas need first to be harvested. Processes that promote free ranging discussion, tease out new ideas, and tackle concerns should be promoted. This will also run to:
It is important that participants be facilitated in tendering their own ideas readily so that innovation thrives.
There is the larger challenge of finding balance between efforts to facilitate diversity versus efforts to achieve work goals. Efforts to be inclusive include a degree of tutorial assistance for new participants. They also include some tolerance for participants who are less efficient at doing the work. Further, not everyone is capable of being constructive and the burdens of accommodating such folk can easily become onerous.
As an example, there can be tradeoffs with meeting agendas. There is common push-back on having working group meetings be a succession of presentations. For good efficiency participants want to have just enough presentation to frame a question, and then spend face-to-face time in discussion. However "just enough presentation" does not leave much room for tutorial commentary to aid those new to the effort. Meeting time is always too short, and the primary requirement is to achieve forward progress.
The IETF's track record for making its technical documents openly available is notably superb, as is its official policy of open participation in mailing lists and meetings. Its track record with management and process documentation is more varied, partly because these cover overhead functions, rather than being in the main line of IETF work and, therefore, expertise. So they do not always get diligent attention. Factors include the inherent challenges in doing management by engineers, as well as challenges in making management and process documents usable for non-experts and non-native English speakers.
On the surface, the IETF's track record for open access and engagement therefore looks astonishingly good, since there is no "membership", and anyone is permitted to join IETF mailing lists and attend IETF meetings. Indeed, for those with good funding, time for travel, and skills at figuring out the IETF culture, the record really does qualify as excellent.
However very real challenges exist for those who have funding, logistics or language limitations. In particular, these impede attendance at meetings. Another challenge is for those from more polite cultures who are alienated by the style of aggressive debate that is popular in the IETF.
For any one participant, some other participant's contributions might be considered problematic, possibly having little or no value. Worse, some contributions are in a style that excites a personal, negative reaction.
The manner chosen for responding to such contributions dramatically affects group productivity. Attacking the speaker's style or motives or credentials is not useful, and primarily serves to distract discussion from matters of substance. In the face of such challenges and among the many possible ways to pursue constructive exchange, guidance includes:
The essential point here is that the way to have a constructive exchange about substance is to focus on the substance. The way to avoid getting distracted is to ignore whatever is personal and irrelevant to the substance.
Sometimes problematic participants cannot reasonably be ignored. Their behavior is too disruptive, too offensive or too damaging to group exchange. Any of us might have a moment of excess, but when the behavior is too extreme or represents a pattern, it warrants intervention.
A common view is that this should be pursued personally, but for such cases, it rarely has much effect. This is where IETF management intervention is required. The IETF now has a reasonably rich set of policies concerning problematic behavior. So the requirement is merely to exercise the policies diligently. Depending on the details, the working group chair, mailing list moderator, Ombudperson or perhaps IETF Chair is the appropriate person to contact.[MlLists],[Anti-Harass]
The challenge, here, is for both management and the rest of the community to collaborate in communicating that harassment and bullying will not be tolerated. The formal policies make that declaration, but they have no meaning unless they are enforced.
Abusive behavior is easily extinguished. All it takes is community resolve.
The security of the IETF's role in the Internet community depends upon its credibility as an open and productive venue for collaborative development of technical documents. There is strong potential benefit to technical documents through an increase in rigor arising from more diverse scrutiny. The potential for future legal liability in the various jurisdictions within which the IETF operates also indicates a need to act to reinforce behavioral policies with specific attention to workplace safety.
| [Anti-Harass] | IETF, , "Anti-Harassment Policy", 2013. |
| [MlLists] | IESG, , "Guidance on the Moderation of IETF Working Group Mailing Lists" |
This draft was prompted by the organizational change, signaled with the IESG's adoption of an anti-harassment policy for the IETF, and a number of follow-on activities and discussions that ensued. A few individuals have offered thoughtful comments, during private discussions.
Comments on the original draft were provided by John Border and SM (Subramanian Moonesamy).