CHAOSS CHAOSScon EU 2020, 23 Feb 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: CHA-IO-SS - Community Health Analytics for Inner and Open Source Software | Daniel Izquierdo

Description

CHA-IO-SS - Community Health Analytics for Inner and Open Source Software

CHAOSS has been on the ground for the last couple of years with a strong focus on open source. Indeed at the very beginning there was some discussion to have included in the mission and vision of the community the term InnerSource that was lately removed. What if we think for a while about InnerSource in the terms of metrics? As active member of the InnerSource Commons, it is worth mentioning a list of them and how behaviours that are assumed by default in open source communities (collaboration, transparency, communication) do not even exist at the InnerSource level. This talk aims at fostering some discussion internally at CHAOSS with the purpose of understanding InnerSource requirements and how with the same infrastructure, metrics and discussions we're having, CHAOSS may become at some point some standard in the industry about software development metrics (and not only open source software development metrics).

Slides: https://chaoss.github.io/website/CHAOSScon/2020EU/slides/cha-io-ss.pdf

-----
Daniel Izquierdo
Director of Consulting - Bitergia
@dizquierdo

Daniel Izquierdo Cortazar is a cofounder at Bitergia, a company that provides software analytics for open source and internal ecosystems, where he’s focused on the quality of the data, research of new metrics, analysis, and studies of interest for Bitergia customers via data mining and processing. Daniel holds a PhD in free software engineering from the Universidad Rey Juan Carlos in Madrid, where his research focused on the analysis of buggy developer activity patterns in the Mozilla community.

-----
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

B

A

So you would like to introduce the topic of cleaner stores. Today, I, don't know if you know Altena source some.

B

A

Ok, so we'll see what this means later but insert is about bringing lessons learn in the open source world into not questions. So this is not open source.

A

It's about learning how to do open source right. We are learning the skills we are learning about, how to collaborate communicate, how to be transparent. So this is inner search. So the talk of today is about having some kind of higher level talk in this case, not that focus and metrics. Perhaps in some use cases, so we go will be flying through some news cases and it's about bringing the discussion of. Does it make sense if we mix this open source, metrics, health and so on with inner source.

A

So we'll see from my perspective, it may make sense that the level of metrics, perhaps at the level of goals, we are not following the same goals but at the level of metrics and to lean people and large corporations. They are using github enterprise fitler. We are using the laplacian stack or some internal infrastructure, so it happens that from chaos means that the tools we already have. We can measure this in large corporation.

A

So well maybe it makes sense, so you will see at the end a bit about me, so I will use this three main URLs here, which is chaos in a source materia where, where I work cements in the problem of large corporations, nowadays is sorry about that things. Complaints about this, so we have silos of developers. So this means that we are producing once and again the same or a really similar piece of software, this country in the next country and in the other country.

A

So if you think about this, it means that we are not talking to others right. So if we look out of our corporation, who else is working in this way in this graphically distributed way, open-source communities right, so we can think of any open source project that they are basically working in this way. So they know how to do this, so this is this is the this is one of the goals of energy and.

C

Specifically, as.

A

I'm really pro open-source. My goal in life would be to bring all of these skills into large corporations, because this means that we are just a click away to produce open source, because we all will understand what open source means and how to work. So we don't need to go through the process of legal marketing, security, things etc. Right. We all understand this, so this is inner source.

A

So, if we think about the basics of open source, this is it right. So we are talking about collaboration, transparency, community, etc, but large corporations are nowhere near from this, so they don't understand this. They need to learn about this right, so the discussion for chaos is what, if we try to analyze the basics of all of this, how do we measure transparency? How do we measure collaboration, Halloween, Mercer community, so we've been, we've been doing this kind of analysis in somehow, but not that in deep.

A

So that's perhaps the main goal of this presentation, some working groups of the inner source Commons. So we are all aware. So we have patterns community, where we have a bunch of proven solutions for certain problems, which is a pattern, the definition of a bottle, learning path where we are producing specific videos that are available for everyone. So you learn about the inner source. What this means the governance were.

A

We are hopefully having a foundation during this quarter, so we've been going through this for a while, and then there are some ongoing discussions around four metrics discussions, which is main topic in this and chaos, def portals tools and some others right. So this.

B

Is this is more.

A

Was the inner source ecosystem? So this is what the inner source Commons is a group of practitioners, where we meet from time to time twice a year study mass chaos con and then we discuss about all of these topics. So one of the hot topics is definitely metrics.

A

The problem we have oh, this is not give, but it's basic. You know nothing Jon Snow, because we have all the same problem. So we need to measure to know- and we have this problem here in case we have this problem. We are community managers, community database, developer, advocate developer relations and we have the problems in large corporations. So if we need to understand what's going on, we need a first face of awareness so to master to understand. What's going on and then once we have metrics.

A

If we start applying policies, you are the one applying those policies I'm the one measuring. So then we can check if those policies are working or not right. So theses were metrics and use cases can do to help. So the question is chaos and energy source.

A

The way we are working is quite similar. Quite the same as we are doing in chaos, because I'm in part because I'm involved communities, so I am printing. This idea of cold question mr. caproni, the one place that I bring this idea of what we'll call question metric approach to to the other thing.

A

So the goal question metric approach is a method that we follow to understand the proper metrics that we need to follow to measure in all in order to have a proper answer to the business goals, because it happens- and probably you know face this- that when.

B

You are measuring things.

A

You start measuring things for the pressor of measuring things. What you are.

B

Measuring things.

A

Because those are the metrics that you are able to retrieve so the way we try to structure our minds either in ourselves in chaos, is, let's start with the goals that, in this case, for inner source of this ones, once.

B

We have them, then.

A

We define a served, a certain set of questions, we'll see an example later. Ok, so then we can see how this works and then, at the very end we have a bunch of metrics. Some of them might be answering the same question once again and so on so well just to let you.

B

A

In our source, we are trying to guess what this means. It's a 5 and what this means in terms of metrics.

A

Some examples: I try to match this one to one I summit, some advice from some person in the room. So if we think about some of the use cases that we have in open source project or open Shoreham offices, and then we compare this to product development or inner source projects.

A

Well, perhaps we are easily different, a bit different concepts right, but at the end we are measuring the same. So we are trying to measure, for instance, at the level of process which might be the first one we're trying to measure how long it takes to the things clock scissors to code review, something to close a merge to understand the several iterations, the lifecycle right. So this is what we are trying to measure.

A

It happens that perhaps, if we go to corporations, they are using a different domain doctor, but at the end, what we have is.

B

Kind of the same, so we are trying.

A

To measure things that happen by itself and- and it happens again- that we are using kind of the similar infrastructure. So that's the interesting thing here so.

B

We can have like well.

A

Instead of responsiveness here, which might be the time the first time to response for a developer to another developer that is coming to the community, the community for the first time, because that developer is a volunteer which is something really important in a puncher's communities. Then what if we translate this into business units today we have this difference of its at the foreign stations. We have business units that are departments in large organization. So then we can basically analyze the same. We can go to tier and say the first time to answer certain questions.

A

Lifecycle, effectiveness, VM is backlog. Management index is a metric use for to understand the maintenance support. How good we are, which is at the end, the number of closed issues versus the number of those issues that the community or the business unit is able to deal with versus the number of open themes that are coming right. So then you can compare. If this is evolving in the right way, then we can talk about in open source, community engagement, organizational diversity, contributors, growth and then in in a more corporate environment. We can talk about business unit.

A

Engagement for usability adoption right- and this is what we are interested remember- that we want to foster collaboration. We want to foster transparency communication, so this is what we are trying to measure and Surin was before talking about pony elephant and so on. So I'm kind of summarizing this in some in certain risk analysis, because if we have a team working well in, if there's only one team working in one piece of software, then if that themes is moved to another place or the branch in the corporation is removed, what happens with that solver?

A

We have the same problem, not answers right. Territoriality is and analysis we've been doing for a while, which is what's a percentage of files in the source code that are, we modified only by one developer. Then again, this is if highly territorial projects are more risky from that perspective, so we are using terms that are quite similar right in open source than in investors. All of these is already supported in in remote lab. So this is the place where we are. Let's say centralizing everything. Thank you. So you have the ready check.

A

All of them have their specific dashboard, that's it. So this is the summary of everything the example I was I mentioned. Before is so we have some gold, we have some questions and then we propose a specific panel right. So the goal here would be it increased energy source project engagement within the organization. Okay. So then we can split this into several questions. You are specifically apply in certain policies right, so perhaps you say projects engagement, so you are rewarding developers if they are doing the first commit to a different business.

A

You need to reconsider it example, but that's the policy. Then we can check if that's actually working right. Today we have certain questions as we pray at contribute. Contributors interact with general trend of my contributors or WordPad. My contributors are following write or port, a collaboration between the different business units, so this is an example of the of this panel. This is of this questions answer. So this is a lifecycle panel, so I will I will go through it. The bars are active repositories and the red line is the Easter tray.

A

So you can see if this is coming up or down, then what we have here is the some certain procedure is by number of commits. So this is a heat map. You have here all of the repositories: you're analyzing. This is over time and then what you can see is where most of the contributions are taking place.

A

So then we have certain procedures that are quite important and then we can see what I don't know if you seen this at the background of the room, but we see that there are certain repositories that are starting: they have certain life and then basically that life ends. So we have the life cycle. We can see now, basically, in a quick view, what's going on and build a wait, so we have projects that to start it this time and then defeatists and the other way around.

A

So we have crates and never existed at that leave stopping so we can see the life cycle and then these are comets so activity and this allows for so it's the same concept. We have repositories.

B

Here on the left on.

A

The y axis and then on the x axis, we have over time and then we have more the repositories with the highest number of pulse ox or the darker the heat map. Let's say the the green color, the the more outdoors have been configured in there again.

B

We can see repositories for projects or or.

A

We fill it with certain activity and so on all the time, so this is this is an example. Another example about collaboration.

B

So this is an analysis. We've.

A

Been doing with red hat with up and see, specifically, some of you may have already know about this, so each of the dots are developers right and right in the middle. We have the projects. So then this is kubernetes. The bigger star is what we call the jellyfish. So this is the analysis of the system plus open sea. If that was the analysis we were doing in this case, and then we have all of the open sea developer schema right.

A

So if we see like a bubble of this, it means that that developer has been working only in that repository or in that forum. Okay, this is open source. Today we see like there is a collaboration across all of the projects right. So this is what this means all of this means right. So if we have developers with a net to kubernetes and Annette to open sea, this means that all of these projects are contributing to both right and then same here.

A

So we have these three developers that are working in fluent in open sea and in coordinates. So then we can zoom in and say who are they? What are they doing if we.

C

Go to the energy sources.

A

Space then we can say who are they? What are they doing right? So what can we learn from them? They are the ones bridging the community right, so they are the one extending.

A

So it happens that the dinner source- we are all working in isolated ways, so energy sources should be at the very beginning, like isolated, like stars right. The.

B

Goal of energy source is.

A

To become like this, a mess, okay, okay, so this is what we want to do. We want to foster collaboration, so this is and with matrix, having this gold questa metric approach is how we we are following this, of course, in the same way that we have this, we can analyze some other examples. This is part of the same analysis and in this case comparing IBM contributions, direct act and divisions, so the green ones are IBM and the red ones are red hot. So we can see for these.

A

For some analysis we were doing for kubernetes, so conceive contain the DJ at some scenes here and OpenStack in this case, so we were interested in understanding what's going on there. In this case, the Red Hat has more developers in general, but then you can see like in opens OpenStack there's, there's some overlap in the interest in the in the project before that secretary acquisition draining kubernetes, probably aware and then as well, and then some others are one. So these are some analysis that we can.

C

A

Brady's right Pat.

B

A

Okay, so I will try again, so this part I popped, open, stack. The left left side, this mostly right. The right side is more mostly green and then in kubernetes there are. There is a mix of this. The rest of these is mainly right. Sorry about sorry about the color yeah yeah I was not a word yeah.

B

So this is this.

A

Is the talk we be given together and what this is still working in progress?

A

So how would you been participating together so from chaos to Twitter source Commons? What are some corporations that are already using blue martin, la BISE and community members, and so on? We had a couple of presentations at the inner source commands in Baltimore, together with Anna she's there in some interesting discussions around metrics with the for north of patterns. The other way around inner source Commons inner source is bring in kind of new use cases.

A

You can see perhaps two requirements for degree mother lab on this case for loop for Victoria, and the point is that we, we are not focused on analyzing, open source software who I know, but.

B

In any case, there.

A

Are some things that are quite common right, so this is this is where I would like to drive the discussion so if you're interested in energy source I'm the local organizer of this time, so it's happening in Madrid Felicity and it's right after is the week the 14th 16th of April in Madrid, so you can have some more information that energy source comes at all. This is all for today. So questions comments concerns. We have two minutes for questions.

A

C

The company, which is one goal through greater source, but not always the only goal way, for example, versus also often an approach that can be used to attempt to drive adoption of internal services that have been have you totally thought about the metrics around that space, in particular yeah.

A

So the way we are measuring an option in open-source is, we don't have access to certain servers, so we don't know things as perhaps downloads or similar stuff that may be populated metric as well, but in the inner source field we have access to the server. So then we are playing with some more information that we may have. So the inertial comment we've been discussing about reusability and adoption.

A

There are some people that are using certain metrics as if there are other business units participate in this participating, so open issues communicating in the mailing list or in a slack channel. Then we start assuming that they might be interested and perhaps they have adopted because they are kind of a fancy user right. They have open an issue and so on.

A

So in this case what we are doing is so we have a business unit that develop typically at the very beginning, something if there are as many business units around and we can say they have produced the software and times today and they are assuming okay. We are saving off this amount of budget because they are producing the software. So this is the way in this case. There are some people not directly.

A

B

First yeah.

B

Kind of getting the data so that you couldn't get metrics in first place, so they're identifying each contributing business unit, I think in general, there's so much they can do. Why would you recommend that you start the easiest hmm metrics to start getting so.

A

The way I I would suggest to do. This is first have centralized place where you have all of your data, and you know that exists. So this is gonna, be your your caldron I can use that. So it's where you will take basically your foot right, but bones are placing anything try to have. First, this try to have a clear picture of what you are following in the short term and in the medium term, then you will have the answer.

A

So I didn't have the proper answer, typically meaner source, what companies are doing our start in trying to understand adoption in the first case and the in collaboration if they are starting to glory just one of the first things that everyone is doing is to start certain pieces of infrastructure. So then people start working there. So our my policy is about collaboration and someone working. So these are the two main areas. Is it yeah coming up next? Thank you very much.