CHAOSS CHAOSScon EU 2020, 23 Feb 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Diversity & Inclusion Badging Program | Matt & Matt | CHAOSScon EU 2020

Description

Diversity & Inclusion Badging Program

The Bureau of Labor Statistics shows that employees in the field of “computer and mathematical occupations” are 25.6% female (Bureau of Labor Statistics, 2019). A lack of D&I is not only a known problem but also acts as a barrier for potential contributors becoming involved with IT as a lack of D&I will perpetuate complacency (Irwin, 2018), avoidance (Sterling, 2016), and a negative collective conscience (Blincoe et al., 2019). Within IT, open source software exhibits some of the worst D&I. The open source survey run by GitHub states that only 3% of contributors to open source software are female (GitHub, 2017). While open source has made strides in some aspects of D&I, such as event inclusivity (Irwin, 2017) and community engagement to support organizational diversity (Alexander, 2019), work remains. The proposed is aimed at contributing to this positive growth through the development of a D&I badging system. In open source, badges are available for community leaders to publicly display, providing a way for projects to signal their attention to such things as community health and software security. Known examples of software badging systems in open source include the CII Best Practices Badge, Black Duck Security Risk Badge, and the PyPi Python Coverage Badge. We can draw from this badging work to improve D&I in open source. While software badges present solutions for most goals of software development, no badge has been created to support D&I initiatives, and the proposed talk aims to advance these efforts. In this talk, I will discuss complexities associated with D&I metrics and propose a D&I badging system that provides open source communities the ability to make diversity and inclusion within open source projects more transparent.

Slides: https://chaoss.github.io/website/CHAOSScon/2020EU/slides/BadgingforDiversityInclusion.pdf

-----
Matt Germonprez
Professor - University of Nebraska at Omaha
@germ

Matt Germonprez is the Mutual of Omaha Professor of Information Systems in the College of Information Science & Technology at the University of Nebraska at Omaha. He uses qualitative field-studies to research corporate engagement with open communities and the dynamics of design in these engagements. His lines of research have been funded by numerous organizations including the National Science Foundation, the Alfred P. Sloan Foundation, and Mozilla. Matt is the co-founder of the Linux Foundation Community Health Analytics OSS Project (CHAOSS). He has had work accepted at ISR, MISQ, JAIS, JIT, ISJ, I&O, CSCW, OpenSym, Group, HICSS, IEEE Computer, and ACM Interactions. Matt is an active open source community member, having presented design and development work at LinuxCon, the Open Source Summit North America, the Linux Foundation Open Compliance Summit, the Linux Foundation Collaboration Summit, and the Open Source Leadership Summit.

-----
About CHAOSScon

Learn about open source project health metrics and tools used by open source projects, communities, and engineering teams to track and analyze their community work. This conference will provide a venue for discussing open source project health, CHAOSS updates, use cases, and hands-on workshops for developers, community managers, project managers, and anyone interested in measuring open source project health. We will also share insights from the CHAOSS working groups on Diversity and Inclusion, Evolution, Risk, Value, and Common Metrics.

https://chaoss.community/chaosscon-2020-eu/

A

Eight thanks for being here right, alright, so today, I'm gonna talk about something and I actually would like your kind of immediate feedback while I'm talking so not at the end, all right, so it's totally fine I can take it. You don't like what I'm showing in sort of old I can I take it over time, so so I met German Frank I'm on the chaos board, I'm. Also a faculty member at the University of Nebraska Omaha, it's kind of been around the cats project since the start.

A

So it's just really great to see everybody here and all of these really great ideas. There's a lot to take in today my own perspective and think about how this this all fits right and then also I was working with an undergraduate student Matt Snell on this idea of DNI badging, so diversity and inclusion badging- and let me talk a little bit about this again.

A

I- would really really like your reactions to this all right and I mean this in all honesty, so metrics obviously help inform the way we think about complex issues and the list that I have here. Dni value, evolution, risk and common. These are the working groups within the chaos project, all right, and so what that means is that there's, a group of folks dedicated to taking a look at at metrics associated with DNI evolution, value risk. So it's over right. All right. Some of the email follow the goal. Question metric approach, bigger.

A

So every one of these working groups kind of follows this approach right, some of the metrics in these working groups, kind of lend themselves to be put into practice fairly easily. So one of the goals and the caste project is so right. We have the metrics that we put forward right and if you took a look at any of the metrics we have published there. They're marked out there they're documents right, so the documents by themselves are not terribly impactful, so they have to be put into practice in some form or fashion.

A

So if we have a metric on something along value, if we have a metric on dependencies, the metric in a document doesn't do us a whole lot of good. So we've got to get it out into practice to actually see it work right and if we see it work we can actually start thinking about it.

A

We can actually start see the impact it has in sort of somewhere within some of these working groups, some of the metrics that we have kind of lend themselves to be put into practice a little bit more easily than others all right. So, for example, commits right. We can take a look at it commits in a github repo and you've seen tooling, that has been able to extract, commits and put it in to a display, that's great, because it's one of the metrics and we.

B

Can actually kind of see it through.

A

Into it, something that is published or something that's in practice in the wrist working group we have licenses, so we have tooling that we can use very familiar with pathology or scan code. There.

C

A

Of there are a number of tools out there and we can draw on these tools to extract licenses out of out of software packages and display those the license results to a screen right and so again, fairly fairly straightforward to do technical solution, some are harder than others, but um D&I is not that way. So DNA is not diversity, occlusions, not something that we can just look at a repository and write software for and and see it right.

A

Okay, you have any comments, you can feel free to speak up, and so it's not easily direct but but DNI is a is a core working group in the caste project and it's really core to understanding the health of an open source project. So how do we take what is being done in the DNI working group and bring it forward right so again, a metric alone? Is it's not great? We need to bring these things forward. So I spend a lot of time on this.

A

For those of you that know me in the cash project you spend a lot of time on thinking about how the metrics that are produced can actually be put into practice across any working group, and so this one's a bit harder, so I I, run and when I actually I think it's kind of appropriate. It's snow and I learned I would just read the book The, Giver and I learned. If you think about snow, you get colder so for this room just.

B

Think about so perhaps.

A

You get colder so I run I run in the winter I run with my dog and I. Do it to think about things like chaos, things right, that's if anybody else runs or bikes, or does anything like that, it's just to get into a headspace to think about things. So I'm gonna tell you what I, when I have been bringing forward within the cats community as a way to bring the DNA metrics forward all right. So you gotta kind of take a step to the side for a second here. So just bear with me here.

A

While I explain this, so there's called the Journal of open source software so for those of you that are looking in the scientific space, if you know the Journal of open source software, it's a way to publish your software all right, and so the journal opens our source.

B

A

Does anybody know our fund Smith, so our fund Smith was in the github open source program office at the time.

A

I believe that he started the journal of open source software, and the intention here was that if you are an academic and you're publishing software, that you can actually go through a process and actually have that software published and you have a home for it right and there's a DOI there's a way to kind of market and the way that the Journal of open-source software works is that a developer or a team of developers submit their software to jaws all right, good Josh, it's on github and Josh actually takes a look at the software and there's a series of things that they take a look at so forth.

A

I have it on here. You have it readme test code, commenting it's a long laundry list of say, maybe 20 different things, maybe even 30 different things. They take a look at with respect to the software, and so, if you as a contributor, you know looking to get your work published and just if you, as a contributor, are missing something a reviewer.

A

Will actually tell you all right- and it's all done on an issue in github, transparently and in the open right, and the intention here is to kind of have an open, dialog and open a transparent dialogue on you publishing your work in a way to move it forward and improve that work so that it can be published because it meets particular criteria, make sense first. So this is. This is the model right. So this is the model that I was thinking about.

A

um Again, problems were openly discussed and then the open source software is published um so the way that it works. This is actually from jaws. I learned that you can't see this in the back, but what it is is there's a simple web page and you just type the title in the software version and their repository where this thing is located. You know you kind of do your due diligence to get this submitted or to get this all in there and you hit click submit on there.

A

If you actually go to jaws and you just go into their issues, you can actually see everybody who has submitted a piece of software to be reviewed. You can see the number of comments doesn't really matter the number, but you can see that this is open public comment and you make changes. So if I don't like what I'm seeing as a reviewer, you make changes to that software. Alright,.

B

So this is just taking a couple days ago: here's our here's, a piece of software under review.

A

At the moment,.

A

And then, obviously, within each issue, there's open discussion about what is missing and what is not missing and again based on kind of passing this you get published in the Journal of open source software. So you remember, this is about DNI badging, just don't free! This isn't about softer journal, open source software, so denied badging at work. What if we could do the same thing for DNI badging right? So here's where you can tell me I'm completely off the rails or say interesting, all right.

A

So the idea, the idea is in chaos we have within the working group. We have a couple. What are called focus areas alright and within the focus areas, are different, metrics, so speaker, demographics is a metric. Family friendliness is a metric and so on and so forth. Alright, so as an event, you would submit a request for a DNI event badge by saying we are a project that attends to monitoring, speaker, demographics, monitoring, attending demographics, taking a look at a code of conduct for an event and monitoring family friendliness making sense. So far.

A

Yes, so we don't ascribe value to this. So we don't we're. Really just saying show us how you were attending to attendee demographics. Tell us the ways that you're doing this. Do you have it published? Have you done this historically right so tell us how your pertain to this particular metric? We don't tell you that particular you know the pie right. We don't tell you that certain forms of that pie are better or worse than another. We just want to know that you're attending to this issue same with attendee demographics.

A

Again, we don't want to tell you that that the attendee demographics here in this room right now are good or bad. We just want to understand that you're taking a look at the attendee demographics, recording it and presenting that information. Tell us the process by which you're going through show us you're going to conduct, show us your family friendliness support at your event. So these are real metrics that have been put forward in the chaos project, and the idea would be is that as.

B

A

Would submit a request for a DNI event, badge right and we would say great: we accept your request. The review for that badge would occur open and transparently with the people who are reviewing so I. Would we.

D

Would need a reviewer, so they're a bunch.

E

Of problem they're a bunch of issues with us. Yes, you can just be explicit about what you need. My badge yeah.

A

So it would look a little like that, so it'd be something that you would put on your webpage for your conference. It's that then, we've gone through so again the author's submit the we could do this at the project level as well. Now we don't have there aren't a ton of metrics on events and there aren't a ton of metrics at the project side right now, but we're building these right.

A

This is part of what we're doing in chaos, so the request would be to receive this badge through an open and transparent process that is completely recorded out in the open.

A

E

Thoughts on this, yes,.

A

What's the origin for it, so the well to question there's a two different questions: right.

B

So the origin is.

A

That as we produce metrics and chaos, we need to find a way to get those metrics into practice, and so certain metrics lend themselves towards software tooling, and certain metrics do not, and so the work that's being done in DNI is not necessarily a set of metrics that lend themselves towards tooling. So we have to find ways to bring those metrics forward and have meaning in the world. So that's the origin of it. What is the demand for it? I, don't know so to be determined. Yes, a lot of questions Oh two.

E

Questions first, what does long term accountability look like holding to these standards? Have you thought it much about beyond the initial like you get it? Yes, now what.

A

Yes, and no so yes, I, hear your question: how that looks fully fleshed out. You know oh yeah,.

E

The second question is- and maybe maybe it was mentioned and I missed it but like how is how are these defined like who has influence over the criteria for the badges? Also.

A

An another question that we thought about I. Actually, let me wait for both of us on to the next.

D

A

D

It doesn't hold up well.

A

I mean that might be part of the discussion that just looking at it an attendee. A registration list is not enough. You actually have to think about how you would sample the audience, don't understand demographics, doing.

D

Something which is provisional initially and then we accept it. Maybe.

A

Yeah yeah I would.

F

Answer the question about the demand, because this set of metrics not only doesn't lend itself to software to him, but the metric itself that you have. This lends itself to an outward facing kind of announcement because nobody's gonna really care if my time to close PR issues is improving like out in the world, but there are a lot of people who would like I believe would feel safer being a part of an event or part of a community.

F

If there was some sort of standardization, you know and again I hear what you're saying you got to work. A lot of this.

A

There, yes right.

F

But that to me would be a good place to start for Adam and if you could actually work these bugs out. Yes, this.

A

Is a proposal right? This is this is straight a proposal and it's hardly a call for help.

A

C

There other comment: I want to make I mean I, forget the name of it, there's a similar badging program for security that Ella is doing and what I. So what I looked.

B

At that, when I was at the LM, when I was.

C

Talking to different community members, taking my elf hat off I, don't care if you get a badging or not. This is a wonderful checklist for lucky event. So.

E

C

For whatever reason, family friendliness might be cost-prohibitive, innovate you may not be able to do it, but I think having a checklist in and of itself is pretty helpful. These are the things that you want to think about before you can you be down here for your vets, whether.

B

C

A badging or not yes, oh yeah, yeah.

B

I totally agree with that: I think it. We did see their core infrastructure initiative and it was really very helpful to to have a cure focus and to be able to move things forward. Well,.

C

B

I would I would say this is very good for D&I, because you and then you can go to speak to you a minute. No, we actually need to take the seriously we need to put research history. We.

E

B

These steps- and it just helps it- helps women move the boundary or the bar. What we think is normal or what should be done, and it helps us to do that. One yeah.

A

Many, the badge would be a way to kind of certify that you're attending to these issues. Yep other thoughts, yeah.

D

Sure family infrastructure would make sense if you're doing the money from Dave and.

A

That may be very hard for half a day. Yes, I mean there are yes I agree. There I probably can't work through every case, Yeah.

D

Right right now, but I think having a checklist for organizers and saying, isn't: organizer.

E

D

Would be great, that's very fair.

A

At two minutes, so again, why do this? Okay? So here the here the challenges, the round that kind of thoughts, so one of the challenges is: if we do this on github, in an open and transparent way, every badge request requires at least one report see the problem. There's this humans that need to be involved in this.

E

Yeah so kind to a humble promotion here for open source diversity. Sir.

B

So there is a inter.

E

Project community collaboration around open source diversity, open source diversity, org, open source diversity in a free node and in matrix I think that's a community that there's not a specific set of goals in place for that community. But it's a place where there are a lot of mines across different diversity and inclusion efforts and multiple free software communities. I think that would be an excellent place to get more feedback on these ideas and get a wider view of people who aren't in the room right and.

B

C

Know, that's not the only place, but.

E

It's one of the best places I know I just wanted to put an insert for that might be a place to drive it forward. I've been told the stuff.

A

Just like that, but you could just kind of read the others volume, maybe an issue of course, and then time the community will release more metrics over time, so how there's early badging may look different than later magic. So it's just that there's a time component in there so anyway, done like that was questions and everything. Yes, all right. Thank you.