Related to my previous post, here is some information about recent efforts in the computer architecture community toward diversity and inclusion, and against sexual discrimination and harassment, courtesy of my Illinois colleague Sarita Adve.
I would love to hear of similar efforts in other research communities, including my own.
Events in the architecture community, posted on the SIGARCH blog, Computer Architecture Today:
- Data on gender diversity in computer architecture (by Natalie Enright Jerger and Kim Hazelwood), including data on PC chairs, SC memberships, keynotes, and awards at the top four computer architecture conferences: ASPLOS (co-sponsored by ACM SIGARCH, SIGPLAN, SIGOPS), ISCA (co-sponsored by ACM SIGARCH and IEEE TCCA), HPCA (sponsored by IEEE TCCA), and Micro (co-sponsored by ACM SIGMICRO and IEEE TC on Microarchitecture). All conferences show much room for improvement, Micro stands out.
- Public reading of a diversity statement at Micro 2017 (by Margaret Martonosi). Call to action for Micro and its sponsors (SIGMICRO, TCmarch).
- Statement of SIGARCH’s past and continuing efforts to improve diversity, with a commitment to do more (by Sarita Adve).
- Data on diversity and inclusion on conference governance metrics (by Kathryn McKinley), going beyond gender, including academic lineage, institution, and geographic diversity, prompted by Micro’s responses to the above events (e.g., new steering committee but without policies for membership, term limits, or rotation).
- Personal accounts of discrimination and harassment at our conferences (by Kathryn McKinley and anonymous colleagues). Yes, I did not think this happened in my community either.
- Announcement of SIGARCH CARES (by Sarita Adve), a committee of respected and trusted computer architects who will be available to help those who face discrimination and harassment at SIGARCH events.
Initial thoughts on lessons learned and moving forward:
This is a work in progress that will continue to be refined here – your input is very welcome. Kathryn McKinley’s post concludes with many action items for individuals that can make a real difference. Several of the above posts describe issues that need attention in my community, perhaps it is the same in yours. Below, I focus on lessons learned from the process that brought us here and what research organizations such as ACM SIGs and IEEE TCs and other related organizations such as funding agencies can do, moving forward.
- A professional digital space for the community to express matters of concern makes a difference. Publication of the above articles (with a comments section) on our blog helped extend their reach and impact. We hope other communities establish similar meeting places if they don’t already have them. This is a lot of work to do well and kudos to Alvy Lebeck (founding editor), Babak Falsafi (SIGARCH EC member who spearheaded its creation), and the CCC blog (our inspiration).
- Data speaks louder than vague perceptions. The authors of the first post had to do a tremendous amount of work just to gather the data. ACM will now start collecting demographic data as part of the conference registration process. We hope other organizations will push for the same and other communities will engage in data gathering and self-reflection similar to the architecture community.
- A critical mass of committed people makes a difference. One or two names appear on the above articles, but each represents the coming together of many supporters, male and female, senior and junior, without each of whom none of this would be possible.
- Working with large organizations is hard but the needle can be moved one step at a time. ACM has a clear commitment to diversity and many helpful policies, but execution requires “on-the-ground” community involvement which is harder. SIGARCH CARES is a small step – it helps for reporting – but much more needs to be done (e.g., mechanisms to ensure that ACM decisions for remedial actions in response to allegations are enforced by the community level organizations such as SIGs). A lot of work is still needed to establish such mechanisms in a fair and enforceable way by people trained in such matters.
- Working with multiple large organizations is even harder. Conferences are sponsored by multiple professional organizations. They are attended by people employed at universities and companies. The conferences and attendee travel are funded by government and private funding agencies. All of these stakeholders need to establish appropriate structural mechanisms to communicate findings for remediation so it is clear that discrimination and harassment is unacceptable everywhere and has broad consequences in our entire professional community. Much remains to be done to establish such communication mechanisms.
Recent Comments