CHAOSS Evolution Working Group, 7 Nov 2018

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: GMD Weekly Meeting: 11.7.2018

Description

GMD Weekly Meeting: 11.7.2018

A

According thank.

B

You well then, I'm going to summarize just for the record in.

C

B

A

Yes, it's going. Okay,.

B

So does summarizing ascend and me were discussing all the use cases and my conclusion and send you say if you're sure it is that this case is to be a specific to what people submitting the use cases want, and it's not that important if they could be, could be made more general. That's something that that we should be exploring when we are working with general approach.

B

Gold question metric for the focus area in general, so that the specific use case can be, in the end, a particular case of what we need to consider in a marginal way when we are discussing the focus arranger in the specific case of array, for instance, what we can do is to consider what he says, which is basically I want to define experience it configures. This way and I want to focus on this and these questions and convert that into the into a use case.

B

And the important thing is that the way we defined focus area considers this use case as one particular instance of the things that can be done in the focus area. Bad.

C

B

Focus area should be more general and all the things that we are discussing, for instance, about how to define contributions or how to define as potential users or something could be parameters for the famillies of metrics in the are more general focus area right. So, if you are we, we can follow on working with this and again in proposing it mostly and second of moving this for testing. Will it works and if it works, go on, if not in a couple of months, we can revert to an another way of words. Okay, so.

D

B

Okay, so in this case, if you don't.

C

B

Write some comment in the in the in the direction that you agree and I'm going to produce the peer base it on what race it and so that he can.

D

Okay, I, okay,.

B

Thank you so, from the point of view of the procedure, this is going to be the first case where we move from the issue estate. Let's say what, when discussing the case in general, to the PR state, where we are going to produce an actual text and now start to discuss on the details that we write? Okay, so no more general discussion about the use case. Now we can discuss about the specific wording of owl who's right. Okay,.

E

Yes, my internet connection is unstable. Okay,.

B

Well, you can be later than nevermind, then then, according to the.

C

B

To the agenda use cases, what the first thing this is the only one that we have opened right now: I'm, probably going to have a second one with somebody from uber I'm talking to them, but I've been talking for a couple of weeks and I still don't have any specific text. Well now. What I hope to have is one so maybe demanding I.

E

Have some I have a yeast case from a some Community Managers that I'm grading as well? Okay,.

B

Then I suggest that that you start writing the ticket or they start writing the ticket them in the issue as soon as yeah.

E

B

That we can discuss in it. Okay, then, if there is nothing else about these cases, we can move to pull requests, which is the next.

A

Use cases till you write them or do you ask the community managers to write them so I'm.

B

Sorry, it was for him so same going where I think.

E

It depends in my case, I've talked to a lot of different community managers, so I, probably synthesize and write it raised, have a unique case because he's very highly engaged in the community and has been for a while, probably better able to articulate it then kind of know how we're doing now.

B

I think the important thing is to me to have them as much involve as possible. On the other hand, we also need use cases so that people learn how use cases are, and they can think of, is my use case by by example. You know so so.

A

If you summarize the use case and write it up, you verified that it captures they're working correctly by giving it back to them or what's the process there so.

B

I would say that the ideal thing would be while we're discussing the ticket. If you look at race here started with a description, the description can be something that chain or you write, and then you tell to the stakeholders. Did you see like have to do all this comment in the ticket? If you want or tell me and I comment to you so that the important thing is to get the feedback in some cases, people may be ready to provide the feedback directly in the ticket in some other cases.

B

Maybe they talk to you and you can come to that. They cannot write. Some community manager said whatever and and then we can work with the eustachian. From that point of view, I think that when this is working, the communities themselves are going to produce the use cases and they're going to discuss and so on, but but for a while I think we need this forum so.

A

B

Okay, then anything else about these cases.

B

E

My perspective see: there's a vocabulary section: okay,.

B

Yeah, the next one would be the full record, which the only one is, the section on vocabulary and Celotex I'm just proposing to accept it. Of course, that would be convenient thing.

E

I think it's just a start. What's good, okay,.

B

E

B

Okay, thank you.

E

Sorry I'd, like I, said I was explaining issues the beginning, I'd better contents for the last week, so yeah.

B

E

B

Frame happens to me many week, so it's just a matter of just a matter of doing it or when we can yep.

E

Yeah nothing else. The meeting makes me look at stuff.

B

Okay, so then, with respect to issues, I don't have a specific comment, except for what we are talking in the in the ticket to themselves. So for for the the two tickets that seem do you have a sign you work and then, when you want, where you can I'll.

E

Have them all I'll have them done before the next meeting? Okay, perfect.

B

But if you're in the heart or something let me know, I can probably woken them too. So, okay, I know what is the spending in a week in a comprehensive manner with that, unfortunately,.

E

I'm not teaching this semester so as bad as I will be able to get to this for next week.

B

Okay, okay, great and then I have nothing else with issues any other comment or no okay, and the last issue in the agenda is the last item in agenda is talking about the first focus area, so I'm proposing this, because my feeling is that we should start working with this top-down approach and romley. We should select one of those and start walking up on them if more than one people want you to work with them, we can select more than one focus area, but for now we can start with one. So what do you think.

E

Yeah I think start with one and.

E

E

I'm, just looking co-developed and I thought I thought we went, that's not Carter and I went through and fixed the broken link, so I'm just wondering what we did that the links are broken again. We.

B

E

Co-Development there's the goal code, development and markdown file and I.

E

Don't know if we okay.

B

Yeah I see some metrics are not working yeah.

E

And I'm not I, don't remember, but.

B

The thing the support of this was fixed, but maybe not all of it, okay. Well, we can work in that anyway, but my proposal was to start working, as we said top down like people in diversity and inclusion are doing so I start by thinking about the goal or goals and the general description of what we mean by in this case cause development, a safari and.

C

B

Think that the first thing would be.

C

Interrupt there and all the links are broken because the metrics folder was renamed okay,.

B

C

You just need to go through and change all the linkster.

B

C

Metrics folder change, yeah.

B

Now I remember, we could say, propose it the change for the directory, but they forgot about the matrix names and underlings. So sorry about that, that's my food.

E

No, no storage so.

B

I'm, going to open and I need you for fixing this right now so I mean while I was talking about the scope.

B

First thing we should define is whether we want to focus on code development or think about contribution the contributions in general, which is something that a discussion that pop up when we were talking about race use case I mean.

E

I think code develop code development, so I think of contributions in general, I think there is code, development and there's issue management and the third primary kind is this documentation I think in a lot of cases, at least on the projects that I participate in the documentation is actually handled through the code development process and many so I don't know if we need a spirit, a focus area for documentation, I, don't know if we don't either. What do you think.

B

I think we can either focus on something more specific and a stick to code or try to be a bit more general and try to deal with contributions. Contributions from the point of view of the check contributions to the artifacts produced and I'm, not talking about reporting backs, for instance, which is also contribution from the point of view of the project. But it's not really a contribution to the artifacts, so you could say oh code or documentation or images or thickness or stuff like that. Right, I would.

E

Be inclined at I would be inclined to focus on code development more narrowly only because it gives us a smaller space to work in and makes it.

E

It just closed this off a bunch of the kinds of questions that I had in the case of raise use case and says this is about code and- and there are a lot of metrics just around code- that we should talk about and we don't need then, and once we have those defined I think it's actually easier to start talking about the broader definition of contribution and how issues and documentation or other forms of contribution might tie into that. Mm-Hmm.

B

E

That makes sense to you, yeah.

B

That makes sense, so, basically, your proposal is to be more specific right. Yeah.

E

Code is code, just focus on that first yeah.

B

Looks good: what do you think I mean the rest of the people? That's Kevin.

E

And gierek I.

A

E

A

Hanging fruits first and then expanding from a working example.

C

Also agree at a start somewhere.

B

Okay, so in that case, what we can do, if you want is I, can open an issue about defining the scope which, basically, we are going to say this and then defining the goals for defining the goals. I would I would say that we can discuss on the ticket, as we did for the race use case, but it would be.

B

We want to understand contributions so what kind general roles we have with respect to that, so we kind of specific goals when we say learning about contributions right, that's what we do, but we define the general goals and then based on that, we are going to defend the questions right so right now, if you look at at the file focus areas called development.

B

The goal is define it only one with one one sentence and if I, how effective the community at merging new code into the code base. So the idea would be to first of all see whether that goal captures everything that we want to know. I would code development or maybe there are some other things, because this is taken, for instance, on how effective the communities- and there are other things and second try to define the.

E

Issues that have come up in three.

B

E

Two constructs that have come up in discussions: I've had around code are velocity that.

C

E

These these are so those are maybe concepts that we want mm-hmm and, of course, as we learned when we're talking to very different community managers, are going to evaluate those constructs are concepts differently. Mm-Hmm.

C

E

Look at commits some may look at lines of code. Some may look at a synthesis of those pull requests. Sometimes Mattern sometimes I think those.

D

Are mm-hmm-hmm.

E

So I think, but I think any code thing that we want to look at is gonna. Have the dimensions of trajectory and velocity that incorporate mm-hmm.

B

E

Is a or parameter so.

B

Another related topic could be quality, some people are interested in quality or at least mechanisms for having quality I mean, for instance, deciding whether at least I have code, review or I have testing or stuff like.

D

E

Yeah I think quality is definitely an issue I, it's it's harder to define because it's more subjective and project-specific and there's there's more debate.

B

But maybe I agree qualities. We are, since we are talking mainly talking about processes and communities here, maybe we can focus on whether the priorities using processes for improving quality, like other using code review for all commits or like I, they have in testing for every commit or every change that they have. They have a corresponding test, stop later, which you know our processes that help to have better quality, and you are not really saying this has better quality, but it says this print has processes for improving quality. So what do you think.

E

B

Or maybe we can discuss in the issue what no.

E

No I think I think I think we do I think the process is for quality are good way to look at it and the presence or absence of code reviews and the coverage of code reviews are those are those are things that can be measured and their process measures like you said, but I think in most projects that I've been a part of an open source community that has the these code. Reviews happen in they often don't happen. Frankly,.

D

E

When they, when they do, who it I think a good thing, but they often don't happen so simply almost it's not binary, but almost the binary of is there.

B

And the same happens, the same happens with testing, for instance. In some cases this you know they have where up this cooperates or not, but they still have been some metrics about that. Even if there are sort of binary could be interesting. I think.

E

Yeah I see testing in addition to code reviews as being another. So if a project I mean that's another, having testing is a see. You know what quality mmm-hmm, I, seeked I, see code reviews recorded on project like we don't have like the otter is pretty typical things. I participate in. We we don't do code reviews per se, but we've written a lot tests.

B

So, if, if you want what they can, what we can do is I open the ticket. Yes calling it focus.

C

Right in and I set up with these.

B

Things that we are commenting now, but please, okay,.

C

B

Anything you may want, and what we can do is to disguise there for a week or maybe two if needed, we can send an email to the mailing list saying we are discussing this specific area in this issue. Candor, if you have something to say and.

D

B

Can summarize, after one of two weeks, we can summarize and add that to the code development era file, let's.

D

B

Didn't define the goals and from there we can go on refining to the questions and, finally, the metrics yeah, which I think it's basically what we are trying to do in the diversity and inclusion group on York you're more familiar with. That is that right.

A

Sorry, what was a exact question, whether.

B

This process of a starting by trying to define the goals carefully and then go to the questions and then go to the metrics, is basically what you are trying to do in the diversity and inclusion working good. Yes,.

A

We started with goals and refine it all the way down and right now we are at the leaves of the process where we defined the details for the metrics and so on. So.

B

This would put us in the same line of the diversity class in question two. So let's start with the goals and see what happens with this code development thing, and then we can expand to other focus areas: okay, yeah.

D

B

I'm on opening the ticket about that as soon as possible,.

B

And that's all for my site, I think and.

E

I, don't have anything to add, so we can conclude this. This recording with our usual end of meeting song.

E

Which gr usually sings.

E

Good everybody, ok.