CHAOSS Evolution Working Group, 7 Jul 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: CHAOSS Evolution Working Group 7-7-21

Description

No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).

A

Okay and welcome to the evolution meeting on july 7th the day after the milwaukee bucks lost their one and only game in the nba finals to the phoenix suns you give chris paul six days. Rest. The man is gonna perform.

A

However, he won't have six days rest tomorrow and we should see a significant drop off in that aging player's performance.

A

I'm a bucks fan um my my giannios giannis jersey arrives tomorrow, along with seven green t-shirts and my cream abdul jabar jersey arrives on saturday.

A

This is the first finals of my fan career for the bucks. So I am, I actually go a little crazy yeah. I actually owned no bucks gear at all, and uh I put an end to that now that we got that recorded important chaos business. I have no clue. What are you talking about basketball, nba, national basketball, association, yeah? You don't yet have one of those teams in nebraska, but I mean omaha is primed for some major league sports team to show up there I mean you guys are over a million people. Now. Aren't you.

B

uh In the in the metro area yeah, I.

A

B

Around 500.

A

If oklahoma city can have an nba franchise, I'm pretty sure omaha is going to get something.

B

Yeah yeah and the uh the fan base here is is rabid right, like.

C

Everything everything.

B

That comes in is generally supported, uh yeah, so I mean we do we do we do millions and millions of people for the uh like the college world series right.

A

Yeah yeah, if, if there was a professional racquetball league, warren buffett would own a team there, because I don't know if he had I'm a big warren buffett fan he added a racquetball court onto his house.

A

I mean he lives in a modest neighborhood, as you know, but it does. He does a racquetball course in racquetball for it in his home.

B

A

It's not that modest.

A

I mean it's certainly more modest than like one of the richest.

B

People on earth would you would expect yeah, he doesn't have a yeah, he doesn't have a crazy mansion, but that, but that neighborhood is uh it's.

D

A nice neighborhood again, four million dollar.

B

Houses yeah, fair and.

D

In omaha, that's on the high end yeah. Yes, although.

A

um All real estate has gone crazy this spring in the united states. As you know so, kevin you were discussing before I said: hey, let's hit record um was that um before I said, hey we hit record where the okay I had the I had the notes up.

B

uh So before you hit record, we were talking about the uh the pull request that we just merged, which officially sends the uh uh contribution attribution metric into community review. So, in addition to that uh metric being merged, uh we also created the uh the community review issue yep. So that is there. uh The only thing stopping us right now is that, what's.

A

On this release,.

B

Is the website which which.

A

Is fine, I mean it's yeah. We have.

B

A

That's all right, I mean we have a formal release coming up at the end, tell me I know we made some adjustments. Is it the end of july that things go under review for no.

B

I believe it's.

A

B

They go into review in august.

A

B

Oh yeah, okay, making.

A

The release releases october- oh okay, all right, I didn't realize yeah. I I.

B

That makes sense uh so the 30 day the official 30-day review starts in september and then the release happens in october.

C

All right, so as long as the website.

B

Freeze ends before september yeah I mean I feel like we should be good. I think we're pretty close on that yeah yeah yeah, uh thanks to uh thanks to matt too, for uh okay, too. Yes- and I have to remember his name for his work- the the website freeze can probably be at this point. It's just a matter of kind of buckling down and and getting it done, and we can discuss that. Okay, so.

A

So the pr is in place, um the pull request is in place and we're waiting for the pre-release review period. um Obviously, okay, so uh pr is, did we create the issue or not.

E

Yes, we know the issue is the issue number four 410.

A

So do we should we send an email out to the community asking for that? You know with a link to the proposed new metric and the issue and just start the review process. I mean it doesn't need to be on the website right.

B

Oh, no, it doesn't need to be we can. We can do that without it. uh Generally speaking, the uh working groups tend to not do that for the uh uh for the continuous release process yeah. I.

F

Think once it's been done.

A

In the past, it's usually because I've done it. I think we should like. I don't know how anyone else is going to know, there's a new metric to review if we don't put it in an email.

B

So when we, when we enter the official review process, we do send out that notification. That's the the right right, but I think I think interim metrics warrant an email yeah. That's that's that's to the community. Otherwise, there's just letting you know that it's not a it's, not an established norm at this time.

B

uh However, I could we could add it to the process. There's a I mean I don't. I mean the handbook. I think how you're supposed to do this and we could add that.

G

I think in the process it is written that after the metric is released and an email to the community, that's what I recalled it was written in the handbook. Okay, so yeah.

A

It was in the most recent newsletter.

A

Because I know we've tried to consolidate a number of things into the newsletter.

F

So that would have been probably maybe this.

F

One sorry I'm just looking at this.

A

Nope, it's not in the most recent one, I'm just going to check the one that occurred. There were two there's been two newsletters since that meeting so.

A

No just meetings.

F

Okay, all right so.

B

Not a bad, it's not a bad idea to. uh We should maybe ask elizabeth if she can add a section for uh continuing these metrics yeah. These are all the metrics that are in the in the continuous release process.

A

Yeah, I think we obviously, since she doesn't go to every group we'd- have to keep that list somewhere yeah. That would.

B

Be maybe we maybe we edit that in the handbook to uh uh there'd be less noise if it went out in the uh the working group or in the in the the newsletter there'd be less noise if it went out in the newsletter, rather than a bunch of emails from working groups, whenever a metric is released.

A

So is this the one okay yep? This is the this is the metric.

F

The metric is here and the issue.

F

F

Probably the top one.

B

We do have some outstanding prs by the way.

F

Yes, I thought we could.

B

Go through those I agree.

A

And this is for what is the? What is the metric formally called contribution? Attribution release contribution.

A

A

Does anybody else want to take the action item of emailing the community.

E

A

E

Nobody else wants.

A

To I can do it that'd be great man. Okay, I am quite grateful to give you that, let you let you take that task, I'm appreciative as well. Do you want to make do we want to have the email make out make any note of these potentially related metrics.

B

uh Can we pull up that metric real quick again does? Do we not? Do we not mention those in there? I guess we mentioned them with the link in that it would be in the issue in this. Generally generally, when I'm, uh when I'm working on a metric, I usually try to include those lengths.

B

The link is in the chat.

B

That's that's something I believe should be part of every metrics definition. That's the how it's related to other. Yes,.

G

We have done it, we have done the linking within the metric if you open the metric or paste it in the chat and like types.

C

G

Contribution and every other there's a like embedded link for that.

G

Yep, so if you scroll down the types of contribution as a link, diversity is the link and all that.

B

So organizational diversity is there? What's the labor investment one? What's that labor investment is from the value? I guess else where, where is this well they're just links so volunteer versus sponsored related to organizational diversity and labor investment.

B

uh Number five is types of contribution which is yep. That's just types of contributions. uh Elephant factor could be added because elephant factory is not there, uh so we have.

G

Not developed the metric for the elephant factor have we yeah? It's released.

B

Like two years ago, so we could.

E

B

Could add elephant factor to it? However, it's not I mean it's, I don't. I think it's pretty thinly related to elephant factor, because yeah, it's related the same way that organizational diversity is related right, yeah, but I mean yeah.

A

Yeah you're right.

B

Which is just it's, uh it's not a strong connection. I would include it just because we talk about it a little bit. uh Okay, because organization is an important uh part of understanding who these people are.

C

B

But but I don't think it's yeah if it if it went uh if it got published without adding it, it wouldn't be a huge deal yeah. That said, I do think we should add it. So maybe maybe we should make a note in the issue to add.

A

No uh okay um yeah! You can do.

B

That is it just elephant factor. Is that the only one we're missing there were?

B

Oh types of contribution uh who is working on the mia who is working on the project would be contributors, so we want to add contributors and as a common metric as a is that a common metric, I don't know which one contributors is, that a common metric uh metric contributors is a who are the contributors to a project uh yeah. This is related so that that bullet point or the in the in the metric itself.

B

The. I think that both point one, we probably need to link that to contributors. Okay, which is in, is that common or is it.

H

ah Hold on I'm, on the I mean.

B

A

The link to the website, so I have the is elephant factor, a risk metric.

B

Or uh I think it's a risk metric yeah, that is a common metric uh yep, so contributors is a common metric.

B

uh Yeah contributors contributors should be added at line one or not line one, but one uh elephant factor should be added at line.

G

So we have already linked that type of contributor in the common metric. There's no uh oh types.

A

Of types of yes types of contributions, contributors is, the fact is, is my comment on the issue sufficient or uh this metric is possibly related.

B

To contributors and elephant factor.

G

B

uh I think explicitly say that uh the metrics should be edited to uh reflect that it's related that uh contributors, all right, yeah and elephant factor- are uh our considerations so not related to. I think these metrics are considerations for the contribution attribution.

A

Say that again, these.

B

Other metrics are related to. I don't think related to is the word. I think it's more of these. These metrics should be considered as part of as part of this other metric.

A

Contributors, I see elephant factor.

B

I guess so elephant factor is probably related. Contributors is, is part of it. So types of contributors are part of it as well, and organizational diversity is maybe related, maybe part of it.

F

B

F

Depends elephant factor.

F

Related metric.

B

Yeah, so I think in the metric itself.

B

Oh where's, that at in the metric itself, in the objectives bullet point one who is working on the project that should link to contributors.

B

What types of contributions that so that already links two types of contributions and then six the objectives at the end, we have. How does.

A

This work, who's.

B

Working on the! What? What's that who's working on the pri on the.

A

B

A

G

B

Line, one of objectives for this metric is to understand who is working on the project. Okay, and that is very.

A

B

The common metric contributors got it and then you're saying sorry kevin. I keep interrupting you, I'm just trying to type this as we go. Oh that that's all right. So in in line six line six we say how diverse is the community of contributors working on a project and that one appears to link to uh nope that links to diversity and inclusion?

B

Okay. So, oh! No, I'm sorry! It's in the filters.

B

uh One of the filters is volunteer versus sponsored activity and within that we say that this metric is related to organizational diversity and labor investment, and I think what we say instead is this metric. This is volunteer versus sponsored, is related to organizational diversity, elephant factor and labor investment.

A

Labor investment is value working group.

F

A

And organizational diversity is that a dna? Is it a diversity? Is it a dei working group, one.

G

B

Labor investment's gonna be value yeah, but what about organizational diversity? That's common.

B

Yep, it's common under people. If I remember correctly, the common working group may have actually been created just to handle the organizational diversity metric. That was certainly one of the origins.

A

Of it, yes, yeah absolutely.

B

True because dei did not want to, uh they didn't want to take it on right, it's understandable, so it's it's slightly different. So.

A

Okay, all right, so I think that covers this right.

A

Yep yeah, I think so all right uh check out the pull requests. We're still gonna work on this reactions, one, but that's old, fixed typos. What is the? What is the reactions fund? This is um emojis metric you're wanted a parameter for this and I don't. I agree with it completely.

A

So I think, but I think we need to take a look at what his changes are. Okay, but.

B

This is, this is very specifically the plus one parameter right. No, this is not emoji reaction.

A

Right, no, it's number of thumbs up or other reactions, because there's a whole lot of other reactions.

B

But the the they mean different things and they mean different things in different contexts. So are we just counting emojis? Is this just.

B

I think emoji reactions are of value. I think emoji reactions have value if there's meaning behind them, but if there's, if, if we don't understand the meaning, I don't I don't understand how 10 emojis in this pr gives us any valuable information. 10 thumbs up in in in a pr uh tells us something interesting.

G

In general, who knows what that means.

G

A

I mean there's, there's uh I mean I think so I would prefer. I like different emojis mean different things, and there is certainly an emoji contribution, but, like so like here, they are giving me github offers me eight emoji thumbs up thumbs down pretty clear smiles, clear party's, clear eyes are I'm keeping an eye on it? Rocket is like to me that would be like awesome.

B

Yeah, I think rock rocket means.

A

Okay, so we don't know sad means or confused. Okay. This means rocket. This means, I I mean, there's only eight emoji and I think.

B

So different projects may use those in different contexts as well, but there are, there are eight emojis and each of those eight emojis mean a different thing. However, this is just for github uh yeah. Are we? What happens?

B

Is this just a filter when we're talking about github or are we talking about slack communication or so? I think I think the communication in general.

A

I think the way georg phrased it is solid. He called it a reaction um and I think probably we should mention. The reaction is the emoji and um really, we should probably show like a screenshot like this so david. So damage refers to them as signals of attention they are. They are that um uh others call them emoji.

A

I mean they're well, the most widely emoji is the most widely known word or used word for these things. um Reaction is github's word, but it's probably different on gitlab.

B

Oh yeah, you did actually you had the comment there where you say signals of attention. Yep. I didn't see that.

A

uh Myth I so we don't have. We can't act on this one right now. It has to be redone to the current.

B

I think I think it needs a lot of work yeah, because, if we're just, uh if we're just capturing, I mean all.

A

B

Add these these lines to these metrics yeah, if we just can, if we're just.

A

Capturing account, I don't think it's helpful at all, so I think I think it's being so. All of these. I think it was being captured for each individual issue or full request, and I mean I how people analyze it isn't really what we specify I mean period of time. Okay, then there then it could be an aggregator.

A

I mean it's an aggregator. So essentially, yes, account. You know like in the past week. You can look at it at the repo level or you can look at it at the individual pull request level. You can use it to draw your attention to different pull requests. Yes, you if.

B

So I think it works if we create if we create and define a metric called reactions or signals of attention and within that metric we we provide more information if we're just adding it to everything here. As a line that's not connected to a metric, then all it is is a count of emojis and it doesn't provide any useful information to any of these metrics you're right. It needs a.

D

B

D

Needs feedback.

E

All right, so um I would say.

G

Plus one for creating a new metric as a reaction or.

E

Emojis or something so, okay, we can alright dispense with this metrics tracking spreadsheet.

A

Evolution see if reaction is already on here, reactions and emoji. All right looks like uh at some point. We already had this discussion yep.

A

So it's considering it goes from considering the under these are uh yeah. Are they sequential.

B

A

So go from now a filter sending.

B

Ready to be worked on so if we want, if we agree to do it, then I think we were depending.

A

So yeah I mean, I don't think it's a priority. I think, can stay where it is right now I, but I think we can close this pull request and say that um we've at.

G

Considering new uh metric for.

I

B

Yeah I do, I do feel like I've argued about this metric before so it's been, it's been a while I'm stubborn.

A

Sorry I mean I'm not fair, it's just easier uh all right. So I'm closing this with comment.

A

That way, it's just gone: fixed, typos keyword and issue resolution duration from filters.

A

12 files changed.

A

Just scanning this, it seems to me there's some potentially significant change in meaning here yeah. That's the comment I made as well.

B

Yeah uh specifically around this proxy dimension change.

B

Which number? Where is that? Which file.

A

uh Down branches.

B

A

B

Yeah, so although code changes can be a proxy to the coding activity of a project change to, although code changes can be a dimension to the coding activity of the project, so it completely changes the meaning of that sentence, and- and I agree that both are possible- uh I mean both actually kind of work in that sentence, but it changes the meaning in a significant way, all right, so the yeah and I've. Actually, I've already commented on that in the uh in the pr I'm just doing a formal. Have you made a video review?

B

F

All right and that's in the code changes line.

C

All right, hello, sean hello, kevin hi, hello, everyone.

A

Hey how's it going.

F

Changes is he this person? Have it anymore.

F

F

Change request accepted.

F

A

I don't think the change request, acceptance ratio changes make anything more clear.

B

Time period, granularity yeah that should actually that change in ratio of a period of time. I think I think they're correct on that one. I think that it should be its own bullet point.

B

F

A

Right I can buy that.

F

A

And the optional visualizations I can go yeah that shouldn't have been left there. Whoops. That's probably my fault.

F

uh Change requests.

F

A

And change requests.

B

So the the idea behind change, requests being a proxy for activity is that the the primary activity in any any code based project is around pull requests right, yeah. So so so, if we're looking at how live a community is how active a community is. The first thing that we're going to look at is pull requests and that's that's kind of that's an important idea for us right yeah.

B

I think so so because if that's the case, then I think we should keep proxy because we're we are making we're making a pretty strong statement there by using proxy right in which range requests are a proxy for activity. They are.

A

That's why people look at them? I mean they're proxy.

B

For some other things as well, um change requests are also just one thing that we can look at right, inactivity which would be the the edit that this person put in changing it to dimension changes the meaning to that right. Where change requests are one of the things that we can look at when we're looking at activity.

B

uh However, what we're saying is change. Requests are the most important thing to look at, maybe when you say.

A

B

There's more than one proxy like proxy, isn't like in singular yeah. I I I agree, but I think the the implication is that because we don't we don't use it in other places right. We use it for change, requests for where we see it. We we don't see issues like for issue. When we define the issues metrics, we don't say issues or activity, or we don't say, issues are a proxy for activity in a project because we only really use it for change, requests.

G

So if you look at the definition they have provided for them, the proxy is like understanding, is more act on behalf of so yeah.

A

I mean I understand: java uses the concept of proxy to represent a software, a software pattern, but the word proxy existed long before computers.

A

I mean lou reed sang about love by proxy back in the 60s, so I have no idea what that actually would be, but it sounds bad. My only.

B

Point is both both words work, however, they mean different things, so the the question is which meaning is more important to us. So I think dimension.

A

Is less clearly less and less.

G

B

If, if this is clear,.

A

Too, it means it doesn't really mean anything.

B

A

Right, let's say.

B

So which it's a weasel word as my master's advisor, would call it if you change that, if you change that sent that word in that sentence to dimension, we might as well just delete that sentence because it doesn't, it doesn't add value yeah, I mean I think I.

A

May have contributed to actually written the word proxy in this metric, but I don't remember but yeah you probably did but proxy. We use that word a lot for, but it means some. I mean it's like really. There are a lot of you know longer. We do this, there's still a lot of people who are only measuring activity. Metrics like and um that's okay, I mean it tells you it does give you a lot of information, but the like number of issues. Number of change requests number of commits code changes.

A

These are all proxies for activity.

G

A

Is it an approximation.

G

Maybe can we replace, like I'm thinking of it, no.

A

Just direct so it's like.

G

A

G

Use proxy as an approximation of something like we don't have that major, but we are using this one or other one.

B

So the informal, the informal is more appropriate in this situation. uh I mean I don't want.

A

This working group has always been like this, where we end up arguing over not arguing but discussing the meaning of specific words and their precision. Well, that's.

B

What um that's, what we.

A

B

We define metrics yeah.

A

Yeah yeah yeah yeah; no, no, it's an important.

B

Part of defining matrix.

A

Yeah and it's like.

B

G

A

Important, oh yeah. I've heard this no I'm, but I think I think, like down to the word. This working group, more than others, has spent lots of time down at the word level. So.

B

I also think that's fair, because I believe that evolution matrix are the basis for all other uh all other chaos, metrics right.

A

B

Is everything is built on top of uh evolutionary.

A

Metrics yeah, I mean we're the foundation, so that I mean we are, I don't know for a foundation, but evolution metrics tend to represent formal definitions of the activity metrics that people have measured in open source for decades.

A

And so that's and I think that's why I like the word proxy, because in fact most of these metrics are simply proxies for activity level.

B

I agree I like proxy as well. uh I just in my comments. I just wanted to make sure that we we considered the meaning behind those two words yeah and whether or not proxy is what is what, as a working group, we wanted to include there I mean.

A

I I think for not. If we wanted his approximation, we could we and and be more inclusive of because really nobody looks at one metric, like change, requests or issues or commits to determine activity level right. They are looking at those like those main three things and, in addition to newcomers and number of contributors, yadda yadda, so it could be. We could say something along the lines of it. This metric contributes to an approximation of a repository's activity level, and then it would be great if we listed the other.

A

I guess I this is, or we could say this metric is one of several in the evolution working group that contributes to an approximation of project activity.

B

Okay, so the approximation and proxy the the meanings behind those two words are different as well. I know I hate to do this to you again, but.

A

That's why I added.

B

A

12 words around approximation.

B

To make it clearer, I think we can, if we just leave it at proxy. I believe that's clear, I don't think we need to. I don't think we need to get into approximation.

B

I see less certainty in the dad's eyes, so approximation is.

I

B

Process all right about a process, whereas proxy is a is an agent or substitute so and in this sentence, what we're saying is that it's an agent or a substitute.

A

Yeah so maybe there's a comment that needs to be added to um the conversation along these lines.

A

Are you I love to review.

G

E

Where I made it, but I wanted to add, do I want to reply to this? Should I you can.

A

E

It don't reply.

I

I

I

I

F

All right: how about.

F

A

A

All right, we are one minute from time, but let's see, if we can finish the let's see, fix typos keyword, proxy okay, we did that one fixed minor issues and metrics for proper displays.

J

A

Talk about that yeah.

J

Go for it. Yeah reaction shoot here, yeah.

A

J

Is actually two small small pr's, and this is about the on ongoing standardization process. Initially, it.

C

J

By josh, but since he is having his exams, so I will be addressing the pr instead. So here it is about just uh adding some new lines between the images and the text, which is because of the new changes that we had in the metric template, which would help us in the pdf because uh yeah. Once again, let me go through that yeah. There is no actual change in the metrics, but just removing some extra spaces and adding some extra lines.

F

Thank you, yeah thank.

F

A

All right, I would just say, if we're going to update the gremore lab part, I should update the auger part so.

F

um I'm going to assign myself to that, so I remember it.

G

I I feel, like all of these things are just space related changes, just reducing this piece, no.

A

I think there's changes to the actual. When I was looking at it. There were changes in the files to.

A

All right, yeah, all right, let's rock and roll, then.

A

All right thanks so much problem um and then template focus, yeah.

J

The second one is, we actually discussed this in our mental and meetings, so the mentor suggested that if the extra data below the focus area is redundant and it is same across so we can remove it to bring it on the same level with other working groups as well, because we are having a standard template for.

A

A

I definitely noticed there was like a lot of extra junk at the bottom of evolution, so I think that's.

C

A

Fix and we're at time, so um I want to remind you, the bucks tip off against the evil, phoenix suns uh at uh 8 p.m. Central central daylight time on thursday, that's tomorrow, so root for the bucks, save your save your soul, and um with that I will conclude this evolution meeting on behalf of all of us. Thank you. Everyone yeah, I'm gonna, stop the recording thanks armstrong thanks, riddick.