It's no secret that women aren't exactly welcomed with open arms in tech spaces. But a recent memo sent around the Internet Engineering Task Force—the open community that develops internet standards—calls out a culture of hostile behaviour and lacking diversity in arrestingly specific terms.
IETF participants Dave Crocker and Narelle Clark penned a Request for Comments document (RFCs are generally how the IETF formally shares ideas) in November that pushes for a change in behaviour and an increase in diversity, and draws a clear link between the two.
"During the early years of the IETF and its antecedent, participation 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," Crocker and Clark write in their introduction.
They go on to describe that interactions within the IETF "can often contain singularly aggressive behavior, often including singularly hostile tone and content," a factor they explain is important because "Many different behaviors can serve to reduce participant diversity or participation diversity."
"It really is about a tendency of some people—not everybody but some people—to use abuse as a control method to reduce discussion"
Clark explained over email that the motivation for the RFC "came from a rising discomfort in a lot of people that good ideas and good analysis were being lost through poor behaviour. That is, you could see highly capable people being put off by the bullying behaviour of others."
Kathleen Moriarty, Security Area Director at the IETF, pointed out that the RFC was published independently and therefore didn't necessarily represent community consensus. She also published a blog post today following the RFC publication, which outlines the diversity efforts within the IETF, such as the establishment of a Diversity Design Team, and claims that bad behaviour within the community is now called out.
"Many of the concerns cited in the Independent Stream RFC7704 stemmed from behavior more common in 2012 and the leadership recognized that action was necessary to improve not only support for diverse participants, but to set a goal of inclusion," she said in an email. "In short, I think we've come a long way since 2012. We do have more work to do and still have some issues, but there is a very quick and open dialog that typically follows any occurrence of inappropriate conduct now."
Examples of the behaviours in question are documented in the memo. It acknowledges that some behaviour observed within the IETF would be "deemed highly unprofessional, or worse" in most professional contexts but has had "long-standing tolerance" there. It reports that excuses are made for parties at fault and that people who do speak out about harassment or hostility can be met with gaslighting: "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."
Crocker explained that a lot of IETF work is conducted over mailing lists with several face-to-face meetings throughout the year, and that the hostile and abusive behaviour could be present both virtually and IRL. "It really is about a tendency of some people—not everybody but some people—to use abuse as a control method to reduce discussion," he said.
"There is in general a desire to behave better, and a challenge of trying to move an entire culture towards that better behaviour"
This is also made clear in the memo, which singles out cases of the kind of behaviour it's referring to that are specific to the tech community. One example is the "vigorous advocacy for a strongly held technical preference." Having a favourite way of doing things is of course fine, but Crocker and Clark note two ways this can be poorly approached by IETF participants: "One is a personal style that is overly aggressive and serves to intimidate, and hence unreasonably gag, those with other views. The other is a group style that prematurely embraces a choice and does not permit a fair hearing for alternatives."
This all ties into diversity because that "group" will often be skewed towards that "American, white, male" demographic.
While these reports sound pretty shocking, laying them out in this way shows that at least some IETF participants are acknowledging that this behaviour is not actually OK. As Crocker explained in a phone interview, the impetus for the memo was an increasing interest within the IETF community to change its ways. "There is in general a desire to behave better, and a challenge of trying to move an entire culture towards that better behaviour," he said.
The detail with which the memo presents these problematic behaviours is intentional: The IETF has an anti-harassment policy and Clark praised the recent appointment of an ombudsperson to help deal with complaints. But as Crocker pointed out, committing to stem harassment and increase diversity isn't the same as actually changing ingrained behaviours. "The goal is to increase awareness of the issue at a deeper level," he said. "It's easy to make a broad statement—and I believe it is an honest statement—that someone doesn't want to be abusive and doesn't want to behave in a hostile way; it's quite another thing to map that down to specific behaviors."
The paper lays out what constitutes harassment and bullying in no uncertain terms in hopes that those broad statements might turn to concrete actions.
At least acknowledging the issue without mincing words helps to bring the real problems to light
It's not just open hostility that's a problem; the memo nods to an informal experiment conducted by a women's group within the IETF called Systers in 2012 that revealed behind-the-scenes biases against women specifically in the organisation. They put forward a large number of female candidates for management positions, but despite the disproportionate number, no women were selected.
"That experiment showed a pretty clear case of systemic bias. And that meant the IETF was missing out on talented, capable leaders," Clark said.
She said that there have been some "brilliant" women in leadership roles both past and present, and that she personally had always felt welcomed and accepted in the IETF, though she noted that she "had the distinct advantage of having been well connected in the first instance."
Moriarty said of the Systers experiment that, "In some cases, this may have been a result of a bias at that time, however all instances of men being selected over women cannot be attributed to this bias." She was in fact nominated for a position at that time, but lost out to an incumbent, which she explained is common in the selection process.
Of course, the IETF is not alone in having a diversity or behavioural problem in the tech sphere. "In the most part the IETF is an exemplary organisation which has achieved a phenomenal amount," Clark insisted. "That doesn't mean it can't be improved and shouldn't set the highest standards for itself."
And in many ways, the IETF is leading by example with this exercise. At least acknowledging the issue without mincing words or glossing over incidents for the sake of PR helps to bring the real problems to light and sends a clear message that change is desired. Clark said she "remained positive about the environment overall."
Crocker was less enthusiastic when I asked how hopeful he was to see improvements. "I vary between being hopeful and being frustrated," he said. "I think that to make really deep changes like this requires active ongoing efforts by IETF management and I think there's less of that than we need."