Jakarta EE TCK, 1 Dec 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: TCK call #4

Description

December 1, 2021 Jakarta EE Platform TCK call #4. Minutes can be viewed via https://drive.google.com/file/d/1ITRLSSPfkZuwJ82hAvEaIV-QdM7znsUq/view?usp=sharing

A

Awesome, okay, so, um let's see uh so first on the agenda, um I see that we have um you're.

B

Not sharing yet.

A

Oh okay, let me let me uh let me do that. Then, let's see share screen.

B

And it has been a while, so you should say this is that today this is the last call of the year. Something cute like this, since we're closing it.

A

How's, that can you see my.

B

A

Too small, let's see.

A

Let's see, oh okay, let me let me stop sharing and let me try that again, let's see.

B

I took a picture and I will.

A

Okay share again: let's see, share and gender sure, okay, it should be just the one browser window now, so it should be bigger and then I can. I can make it bigger as well.

B

That is perfect.

A

Great okay, so um so we're yeah so we're so we we are um um just heading into the heading into the spec api ballot process um and um yeah. So you know some some specs have started to. um um You know um apply for um um the you.

C

Know you know you know.

A

Before the specs, be, you know going through, you know going through and and finding you know getting finalized for ee10 and um for many of the you know, for I think you know, for you know for all all of them. You know that initially um they need tcks generated from the platform tck, which lines into this discussion here um we need to generate standalone tcks, uh for you know, for those and for the other specs that you go into the ballot process and that's a just a necessary um limit.

A

I'm sorry, I see what okay sorry, I got a little confused. I was on the agenda from the last meeting because I had scrolled down sorry about that, um but yeah, so we're um um yeah, um so we're we're we're just getting into the first wave of um of the for the spec ballot process, for you know, annotations, concurrency, messaging and persistence, not necessarily in that order. But these are. These are the specs that don't have dependencies on other specs and they're perfect to do first, so we um I, I think, what we yeah.

A

We need to get documentation, changes in um yeah, we'll want to change from ee 9 to ee10, and you know, um update the you know the spec versions referenced and you know those kinds of things that you kind of expect in the user guide and the funny thing about documentation changes is they tend not to happen until the very until the very end or sometimes until it's too late almost if any contributors are looking for something to work on um documentation is a you know, great area to jump in on, um and we do.

A

We do have a number of pull requests from the spec teams for new tests that they want added for their specs, and you know we'll look to get those emerged in. I know there's a few for persistence and some others as well.

B

God and everyone, I have a question if we have pull requests that need some work, especially in documentation. Why are we not uh submitting collateral tickets to push those messages for call to action to get contributors to help us not only the current contributors, but also new ones that are looking for projects to help? um Can we.

A

Do it yeah, I I I I think it's um a great idea it it it it's it's it's. Basically, um the documentation ends up not being um a high priority. You know compared to um producing the tcks, um but it's important to get done and um yeah yeah.

B

It will help the release vote, it can be and the bullet can be cancelled because of bad code. Documentation code, there's nothing with that documentation. So we can. um We will not make this about media, but definitely with every pr that needs call to action for others.

B

um Let's push it and ask haley that is the one um in charge, one of the people in charge of the jakarta media to you know if we don't, if you don't have the they get each git id, I can give it to you in this call, and then we can say you know this and really quick, simple messages um on how to do it, so they can just push it.

A

So contact mac, oops.

B

I will add the link, so you can see here. Okay,.

A

Okay, great: um um let's see contact marketing to uh recruit.

A

Contributors to help with yeah, I I think, there's a lot of um like old terms, still lurking in the documentation and it'd be good to get someone. You know get get some more hands on that. To get those changes in there and in terms of cancer yeah, we don't have to cancel a ballot. It's just you know. The ballot just doesn't complete until everything is done, but uh but.

C

A

So, let's see uh um so in terms of um staging, you know, you know basically yeah when we're at a point where we think the tcks are uh ready. You know we, you know we, um um you know we'll, expect the spec team to run an implementation against that. You know that particular um tck and um give feedback um on any any any further changes that need to be made documentation and need. You know, changes and create pull requests or whatever you know. What? What?

A

What have you um and um but at some point you know the state, the state that's tck. Will you know you know, will be staged and will be. You know passed by an implementation by the spec team and will um the spec ballot. You know the spec will go final and the tck will be um copied um um to the final sp. You know with the you know to this to the um eclipse you know to the download site and the spec will be updated to link to that.

D

A

That tck and that kind of, like seal seals it off um we're still kind of scrambling to get the tcks ready, though you know we're yeah we're we're getting closer, but um you know there's lots of issues, and I mentioned here: we have 19 open issues with you know: label labeled for ee10, and some of them are being worked on, but many of them, you know you know, aren't done they're, not all necessarily going to block.

A

um You know spec ballots, um but something you know some of them may and um yeah. um That's. um You know at a high yeah at a high level. I wanted to mention that um um I know personally, one of the things I've been working on is updating, updating um the signature test tooling. To make it you know to get it working.

A

Sorry, sorry, um the signature test generation scripting, I should say uh we didn't actually have to change the tooling um to make it easier to generate the signatures and we have a script for generating that and um we're just you know we're just trying you know trying out some um um some of those changes still and looking to get that merged in. um So, that's you know one of the things that the you know, yeah, that's required for you know for these spec ballots to you know to work with the the new signature.

A

You know signature files and we're working with um staged copies of the spec apis.

A

um You know on that um so yeah um um yeah is anyone else want to talk about what they're working on for ee10 on the tck side.

A

A

Hi, cesar sorry, I didn't uh see you waiting to get in no worries. Just uh no.

D

B

A

Welcome um so does anyone want to um someone else, have something to share about what they're doing on the tck side for ee10 that we should add.

C

High uh hi uh scott uh uh hi, uh basically I'm working on uh json, b and json btcks for signature test migrations, uh also working on uh jaf tck, which I have already completed, uh basically uh for the signature uh test upgrade and the upgrade of the signature files.

C

Also uh I'm working on the male tck for pluggability test for the male dck yeah and for uh jax ws tck. I have already worked and raised pr for merger of jws and jax ws test suits so there there is an issue open in jakarta, atc repository for the same.

C

So these are the issues and uh tcks which I'm currently working on.

A

Okay, um yeah on the jax wws, I think yeah. um um We definitely want to merge that this week. I think if um you know, pending input from others, you know it looks like it's approved, so it looks that looks good uh good thanks for the update guru and for the.

C

C

We are, we have already worked on uh um xml binding, uh jakarta, xml, binding tck uh that tck has been built and uh uh issues resolving the new spec of 4.0 has been uh completed.

C

So that's the another update which we worked previously for jakarta. 810.

A

Nice, okay and and it I I I've, seen you working on glass, fish running with glass fish as well, um which is you know, yeah good good. You know good, you know good to see you as well, um and um I I've been trying to um yeah I've been testing locally with with wildfly, and um you know things haven't changed so much that you can't do that.

A

So it's it's good that we, you know we're able to do that still um and, of course you see differences because no one's using the new spec ap, the new specs. Yet, but it's you know if anyone else you know from other teams wants to do that. um It's something that you! You know that um um we yeah. We can point you to.

A

um um You know tck downloads, um um you know to use, and um so you can. You know, give that give that a shot as well, so um good, uh okay. So unless anyone else has anything other kinds of status, yep.

C

Go ahead, yeah, hi scott! uh This is alvin here, hi yeah, uh so uh I was focusing on uh separating uh jax rs, uh which is the carthage successful services tck uh out of the platform tck. So uh we have. We are almost done with this particular task, like uh there are a couple more pr's that is yet to be merged to complete the separation of this tax rstck uh into the uh corresponding api repository, uh so yeah.

C

What is pending in that side is the documentation, the signature test, vr and uh an improvement uh to add a vendor specific descript deployment descriptors. So, apart from that, we have moved, or rather we have migrated, the uh all the jacks rs tck test uh from the platform tck into this uh jack serious api repository nice. uh This means we will be uh stopping the generation of chaksas this one tck uh from our uh the kadhaiti secure repository.

C

uh Also. It also means that we need to run this these tests uh separately uh when we are running the uh platform. Basically, since uh that those tests won't be part of the platform pc anymore. Yes, yeah is.

A

C

A

Sorry is this um going to be done for three one as well.

C

Yes, okay, it is okay, it's going to be done for 3.1 jack, sarah 3.1, so this is one of the tcks that is being separated uh for uh jakarta.

C

Guru mentioned jsonp and jason b. I think most of the work is completed for that also.

A

Great yeah, um okay, terrific, um okay, let's see.

D

Okay, I can go uh quick scott. This is scott.

A

D

Hey so um for catch um we're a little behind with the release uh we not ready to submit the.

D

I forget what release review at the moment, but I think we have the major issues sorted out so looking kind of good heading into next year, I haven't been touching the platform tck, I'm just hoping to get everything squared away in the standalone batch pck and then go delete. All the batch stuff from the platform.

A

I'm glad you bring that up because um yeah yeah, it's it it's I we we haven't, really talked for a while about that and um um uh hasn't been updated. Yet yeah.

D

I kind of I kind of wanted to make sure it all worked first before deleting it and having to add it back in.

A

Yeah, here's he so the I I don't have the um the files in front of me, but um um I think it's I I think um I mean the standalone batch tck is definitely like handling like like jta integration like for running the you know the the the tests under your transaction manager, which is great, um I think the things that the part that might be missing might be um integrating. You know the achillean integration and shrink wrap integration for for uh running on ejb um web and maybe jsp.

A

um You know servlet njsp, um um just going from memory. I I think because I think I think I think.

D

Yeah, I saw a comment I mean I think in the past we go we've done ejb and web going back to uh cts and signature tests and I'm not sure why we'd need to do jsp now, but um that's that's yeah. At the moment the dash cck only has a servlet wrapper. I think, but I think it's, I think, it'd be easy to add the ejb.

A

Oops um yeah the um I I I only mentioned jsp we, you know we should check to see what we're currently what we're currently doing, and um there was some reference that just said web and then somewhere else. I saw it to find that web means gsp and servlet, and I I had to double check what we you know. You know which you know which we ran under before, and what the platform requirements are um and kind of.

A

You know it's something to discuss, I think and and and figure out, um if you, if you whatever they are. If, if you can add the handling of those, um that's great, I think, um if you don't have any, I don't think you've ever had any vendor specific deployment. Descriptors used in the batch test.

C

A

Past so that's perfect um um other! You know for other team.

C

A

That could be um like jack's rs. You know, you know that we were just talking about that. um Vendor deployment descriptors need to be integrated in some fashion and if there's no need for that, then that's yeah that that's great, um that it'll be easier than you. It sounds like you just have straight tests, and you know you know: shrink, wrap and uh yeah.

D

A

Yeah and that stuff and that yeah.

D

Yeah so yeah, so the um test suite proper is defined as a junit, 5 suite or you know, collection of tests, and then we add the artillion extension as kind of a sample way to plug in to say unit five a day unit, five extension, that's also in our chilean extension and that piece of it doesn't look like it. You know fundamentally belongs in batch.

D

It looks like maybe more of a common thing or maybe an artillion or junit thing, but you know at the moment- or you know, to start we'll, just put it in the best pck and worry about where it should really belong later, but that allows us to use. You know the same tck in you know in like an se mode against the sc non-platform implementation, and so I think that's a good way to draw the boundaries and we'll just have a switch to you know, turn on e versus se, profiles: sweet whatever.

A

That's great, that's terrific, okay, um that was that was that yeah that that fills the um the the the the need for the plot handling the platform requirements. So that's great.

D

Yeah, so the one thing so that always good the one thing I was a little less sure about was the signature test, because we were we we had um in the past also documented, even even in the last release of the batch tck standalone ttk with the signature test, and so I you know, I saw the um the new fork based on netbeans and the the maven plug-in around it, and you know just kind of kicking the tires and trying things out.

D

um You know to me: that's this whole area is kind of a bit of a confidence thing. Yeah I mean I don't really understand what the tests are really checking. So you know I look at I just kind of ran through a couple simple tests to confirm that it seemed to be working the same way and you know there were a number of kind of small incompatibilities.

D

um So I guess, if the sense is that we can trust, you know version one five or whatever of this maven plug-in and and we like what it's doing fine, but you know just doing like a dummy check like okay does it handle the last?

D

You know releases signature files, the same way uh it kind of doesn't and and the one dummy check we would use to make sure that we really were doing checking as we would delete the cdi uh implementation or the cdi jar from the class path and would throw an error, and you know that has been downgraded to a warning.

D

um You know I know there's this effort to well, I I don't want to complicate it into the effort of doing the same. I know there's also this thread on the getting a a single file to work across java levels. I saw you yes to that.

D

So I went to look at this project and I noticed there's not even an issues tab in github.

A

Right, that's yeah. You can create a pull request and that's basically it and uh yeah it's. um I noticed that as well yeah, um so it um if there are changes needed. uh I mean I. I agree. I've seen like between the maven plug-in and the ant plug-ins and the.

A

The straight java classes- um I don't, I don't think they're, I don't, I don't think they're equal. I think there are differences between them and um um the, um but I I think we could, probably you know, get the get working within the batch tck. What we you know some similar form to what we have working in the platform tck and if we want to be compatible with the netbean signature test tooling, without megan further netbean, you know signature test tooling. That would be the easiest way to do that.

D

Yeah I mean I really I want this to be an all maven tck standalone tck. So if the sense of the community is that this is a good tool, this is a good fork. Then I'm happy to just go with it and and deal with its idiosyncrasies and make maybe make a few pull requests.

D

So I feel like you're, you are blessing that direction.

A

Yeah definitely yeah using netbeans is definitely a good direction and um um I I don't know for sure you know I. I believe that, like with the maven plug-in, um um it's it's different, but um it's a good. You know it it's a fine direction, it's it's. What others have are using so yeah cdi is using it in bean validation. Tck is using it. So that's fine. Also, if you go that direction, yeah.

D

Okay and I'll try to keep noting these small incompatibilities along the way just so we so we all look like we know what we're doing, and even though this is all kind of so like a crazy format to me. But that sounds good.

B

Scott, I have a question for you. I haven't seen an email from you coming with the work you're doing with the tooling with me when you said anything is stopping you from providing updates on the progress of this via the melanines.

D

um No, I I just was trying to collect my thoughts. I I I just figured I'd start here. Knowing this call was scheduled, but yeah I'll share it. When I come to a checkpoint.

B

And my recommendation is clearly: you can do whatever you want, but my recommendation is to avoid vacuum. Brainstorming because it is a stronger, a tooling is a stronger where you have others to support it. A tool will need maintenance, because otherwise it becomes that, and if you don't.

D

Yeah yeah, and that's also why the fact that there was no issues, even I mean that's usually where you discuss proposed changes to a tool, and I again I was questioning. Is this really what we're all you know the projects we're all focusing on, but I mean you can discuss just as much in full requests. I guess.

A

It's it's a huge area, all on its own and um um and there's very little interest in it. I um I found um other than I mean, there's it. There is there's interest in it, but there's not a lot of interest in the internals is what I've found.

A

But um but I could be wrong uh if any of you all are interested um yeah, and I I apologize if I um said that you're not and you are, um um but just just based on I've, communicated it a lot through issues and um um I did get like feedback from bj, which was great um and some others, but um it just seemed like that. You know um um there wasn't a lot of interest in updating the tooling I should say, but.

B

Let me interject here: let's use the stats, there are only 30, watches watchers and the jakarta ee tck ribo. However, there are 103 jackets in the melonist. If you want to get feedback go to the area where it has the most eyes and brains. That's it simple move forward, so scott move to the mailing list bring the prs or the issues to the you know short and sweet because they get issue.

B

The melon is just to awaken the conversation. The conversation should happen and they get issues or prs hopefully and a little bit and then melanis too, when it's too complex right, like brainstorming, um but we did notice early this month or last month that some things apr was going to be married without any discussion.

B

So it's you know, it just makes good form to always use the open, the forums just because everyone works in different places. Not everyone is interested in getting the repo um information only 30 and we have more than 100. um You know subscribers yeah. So that's a recommendation. I hope you take it well.

A

B

A

Other the other thing I was going to say, um there's a lot of future work. I think that needs to be done in the in the center test tooling, and you know somewhere, um um but that yeah, that's that I think that's out of scope for out of scope for today, but I just wanted to mention that that there's a lot a lot that we can do and to make it simpler and easier for um the future generation or you know our future maintainers to you know to deal with it.

A

That sounds good. I.

D

I just wanted to run it by this conference before opening some issues on a project, and it turns out that you can't.

D

Just making sure I was up to like the latest thought and it sounds like I am so yeah I'll I'll open some tr threads et cetera yeah. That sounds good.

A

You can open issue. I think you can open issues on the open, jdk equivalent signature test tooling, um but the barrier to contribute there is higher, uh so it's easier to contribute to netbeans, but just no issue tracker but but yeah, okay! Well, let's, let's, uh let's go on! uh Is there anything else on batch before we move on.

D

That's not bad.

B

Okay, I have a question about the tck: I'm not a tck, but release 10. This should be done in parallel. Okay, scott! Can you go down to one of the items I brought up here? The second one general? Oh, no, this one, the jakarta tck um thread from november 10th by david blevins states that we, the tck repo, needs to be moving to the jakarta from the ee for jerry people to the jakarta ee repo, and we have six months to complete. It should be done in parallel with the work done and 10.

B

how what are the? What is the plans for the transfer.

A

Okay, I don't, I don't recall the. I don't recall the the um in parallel with 10 release pot, but I may have. I may have filtered that out from the point of view of.

B

Can you open them? Can you please open the link and we can read it really quickly, all right in a reminder to everyone? This was decided by the student committee last june voted in and it never happened because we were received with 9.1, which was a struggle. If most of you remember, okay,.

A

um But but I I I mean I'll I'll, I yeah it's not opening for me right now, but that's fine. um I I certainly believe you know believe you, so how could we do it in parallel to 10? I don't think we could do that without without um suspending um the 10 release, and um if that becomes the plan, then that's fine um yeah.

B

A

But but I think I I think it would be so easy so I'll say either suspend the ee10 release. While no that's not an option, though repos are moved. um You know, you know, you know, you know it as long as it takes to.

A

You know to you, know, to um adjust, um and- um and I don't know if it's like big bang- you just go through and then they're all moved and and and that's it are, um or I think um um you know do it after um ee 10 um completes um you know, release final would be another option. um I don't.

D

Know if there's other.

A

Options, but um it's I think that the the platform tck would just follow whatever you know, whatever um you know it's done for the ee10 release as a whole. um You.

C

A

On this, but uh I don't think we you know yeah, I.

D

A

Really yeah, I don't have much more, you know just to say other other. You know about about that, but I think that would be um you.

C

B

A

Are two options that come to mind? I don't know if that helps, but.

B

Yeah- and I have a question I think the suspension of if 10k is not snl, we must get new features and um all the the nice work done. um I think after works great one of the concerns that I have is that through the mailing list, there have been no follow-up of the plan of attack when we ignore um messages like that. That say you know there has not been a issue created to say, you know, add the link of the request to say this will be done after the you know.

B

We need to schedule it so that it becomes true, so I'm asking we can do it after, and that is great, but we need to create a a get issue at that link. If, if the link is wrong, sorry about that, I was really quickly trying to add that um before this call, but it needs to be a follow-up, you cannot just die in because things that are not seeing every day um get forgotten, and this is very important.

A

Okay, all right, so I will uh let's see.

A

B

You so much brian. I was like no coffee yet so no excuses, though mistakes. Mistakes, apologies, scott,.

A

No problem: uh let's see, okay, all right so yeah I'll um um um yep, so yeah. We can continue to continue that discussion on the mailing list. That's fine!

A

Let's see, let's see, okay, all right, so um um so the other thing you know so the other thing um you're going back up to the the triage role and the platform tck um was interested in hearing from others on the idea that, um like I guess, started on microprofile um as an idea to have um contributors that aren't committers more actively participate in the project and um kind of you know triage implies, you know implies you know a lot, but some you know some of the things they could do they could.

A

They could um basically help manage the project um and you know dealing with you know, issues and pull requests, and you know labels and certain milestones for issues and just you know, kind of helping out- um and you know, on a personal level, um one of the problems that I've experienced with the platform tck or leading the platform tck.

A

um There's with there being so much work, um I tend to have to go, you know, heads down and and help do work um and don't you know, don't necessarily get to um step back and look. You know look around at the overall issues that are pending and to kind of like issue triage and and um that kind of thing as much as much as I'd like to and um I'm thinking you know, I'm I'm thinking um it would be nice to have.

A

You know more help um on that, and you know that's how I was kind of seeing this triage contributor right there. It let's you know, let's more people, um help out in that on on some of these things that we've listed here and I'm kind of curious. If you know anyone has input on some of these options, which I think you know we could take, no action, we could add a you could have.

A

We could try on a trial basis at one triage contributor or we could follow the uh micro profile idea of having multiple contributors and just wanted to open up the floor to input on on those ideas.

B

Who here is not a committer for the ttk.

B

All of you are committers except me. Clearly.

D

I'm not sure if I'm recommending.

B

Okay, so scott, you are not a committer. Who else is not a committer.

A

I think that's it. I think it's okay.

B

A

Everyone else is a committer.

B

So scott you're doing you're going to do some work, you're doing some work in the tck and you're. Creating it issues. um Do you consider it valuable that you would be able to not only assign the git issues you create to yourself, but also maybe others for review? um You know all these things that say here. If we see is cap, can you share the example? I just added on the micro profile check microprofile? No, no there here.

B

No, not that one! That one is there give the shoe. um Oh okay, I'm talking about here. It's an example of how it's being added click in the micro profile. It will bring you to the page and microprofile decided um to to that is the the repo right. uh I think it's the wrong one. It's got. um I I added an example.

B

Should it be well, if you go to the the micropro, the microprofile page under the eclipse infra, I think it's that one it should be yeah and it's not a git issue. It's just like something.

A

No, not under jakarta would be.

B

Microprofile, I want to show an example, and I thought I did it, but let me find it again hold on.

A

B

Yeah micro profile, but it should be under the what's the eclipse, infra micro profile when it shows uh projects. My internet is not allowing.

A

Me uh should I stop sharing, so you can share no.

B

No, you can find someone can find it's not the microprofile way. With page it's a microprofile under the eclipse foundation. Infra that lists the project.

A

Okay, let's see uh oh.

B

Here I found it project eclipse hold on I'm trying my internet because of sorry about this super there.

B

Here it is, I found it sorry about this.

B

Here is cat. This is a link, let's see so you click and I share it in the chat so you're clicking the link.

A

I don't know if I can see the chat. Oh.

C

B

Maybe because you are sharing.

A

B

A

Let me stop sharing for a second and I'll go to chat, yeah.

B

My the zoom is eating all my my bandwidth for some reason: okay,.

A

Okay and now let me um I'll share again and show that screen. Let's see.

A

Oops me share screen.

A

Okay, okay, yep, so you should see it now.

B

So, let's cut go to the bottom and under overview here and the contributors. These are the added trash uh triage um microprofilers. They are usually active in microprofile marketing and also business, and they get added in the wiki that is linked. um The documentation, teach you how to add and remove contributors.

B

If this is active and there is nothing wrong with with trusting new contributors that have been in the project for a while that you know. Sometimes. I believe that um this bug, that is a was a four years old class bug, was revived by me. Actually because we had many people joining microprofile and they felt frustrated, they couldn't do a basic work using github right and then I asked the infra team in the clips I said: do you have anything, a bug that has been complaining about this?

B

Let's say we have a four years old bag, but you can use it or create a new one. I said no, let's use it and we revived this. It took us eight months to to convert the bug into charge, so that means it changed. The entire structure of the eclipse foundation.

B

How get in contributors get to work now every project, no matter what it doesn't need to be under a working group? Can you stretch to collaborate? That means there are no secretaries, it's kind of you're, saying you're the tck lead and you need help. That means that others can come in and it's not that they can merge prs, because a merge only means a committer can do, but they can do all these other things that are a huge plus. I think limiting the access for anyone, they might ask, be a git ticket.

B

Hey say I am a new contributor and I would like to be added. There is nothing that can be done. That is wrong. By having the accent it's just so cool and highly recommend it.

A

Okay, so contributors are that all have the triage these it's looking at the.

B

A

They all helped triage okay, so.

B

All of them have the triage into change into what is cut. For example, you just need heat email to add myself, I requested already, because I wanted to bring the conversation to the tc. um I said: hey me, let's add this. That means let's talk about this.

A

So let me flip this around um does is anyone against adding um contributors uh with triage rule um to the platform tck um and you could yeah? I I guess you can give me input um privately, if you, if you, if, if you wish um as well, um I didn't see much input from others, but I just wanted to open it up in case.

A

um Others are interested, um um so I I so I I think um um I I guess one question: I have amelia if you, if maybe you could help um how do contributors um like you created an issue that kind of kicked this off, which is great. um How do contribute contributed editors find themselves in the like. You know to you know: how do they sign up to get the role? Do they send email? Do they volunteer or do they just get picked.

B

um They get, they send a git issue via whichever repo they work in microprofile and they get added by a committer. Clearly there is a review is like who are you? You cannot add someone that you do not know. The person has to be there for for a little while, right, like you need to trust the person, even though nothing can be broken really, um it is by merits right. This is the first step to becoming a committer right like not on the first step formally to be added right.

B

You create the issue clearly jakarta pck will have to. I would recommend just to copy the git hub that we have a mic of the file. We add um we from oracle. uh He was a pilot and I was a commuter for microprofile. Microprofile has 24 plus repos, but we are one platform, so anyone that gets added gets added to the entire 24 repo, which is so cool and in the side of the in the side of the um what we call it jakarta.

B

Ee tck is very isolated to other projects, so it doesn't mean you are not pertaining to force other projects to be. Like platform or whatever you call it, it is only peculiar to tck. Some projects might choose not to. You need to be very careful so here when we see the gate, the git issue, I will just say you can implant it as a documentation um in the jakarta tck and say: please submit a git issue.

B

If you are working in anything tck and you are not yet a committer, we can add you a stretch and you can have access to do these things simple. You get it and you get the git issue and a committer will have to um use the git, the wiki that we created um to just go into the page of the project under the eclipse foundation.

B

Here the wiki explains to you and then you go there and then you add the email within seconds the the contributor gets added a stretch. They get an email from the from the the tool and it's done it's automatic nice. Okay.

A

All right, thank you. Amelia um great.

B

When and clearly emails are private, so if someone is uncomfortable, then is when a good idea to submit the kid issue and then be a private submit to scott or anything that wishes to say. I have your email. I will do it if you do not just contact the person privately, because we have european laws and we cannot avoid. We must be compliant so fyi to that.

A

um Okay, um so I mean sorry is that so on the eclipse side, the emails price that gets sent as private? Is that that's what you know.

B

I'm talking about, um we have private laws on sharing emails and I don't have a problem sharing my emails. Rarely, but you know if you're living in europe- and you know you prefer- to give to your email privately.

B

That is a great way to contact the contributor after you have submitted the git issue and say this is my email. Thank you so much, because what we did with the feature at the beginning, we added the git issue that is linked to the contributor.

B

Can you click on the wiki it since we have a few more minutes- and this is maybe one of the last topics for today- okay and thank you so much scott for sharing theirs.

B

So if you go down it tells you what back we fixed, and then it tells you you need to sign the contributor agreement clearly without getting true. You are worthless right and then it tells you how to do it right like it goes down.

B

How committers add a contributor now through that one we go a little bit down, is just a basic information about git accounts, but then how committers grant react access to contributors, and in here it tells you exactly what to do?

B

You go to the projects management page under the eclipse foundation. You click edit. It is super simple. I.

A

Got you okay? So if I look at the issue, go back to look at the issue you created, then for a second.

A

B

Let's see- and I didn't add my email because you have it- I took it for granted. I.

A

I do have it, um um but, but if, if you, if, if I didn't have it you, you would have maybe have entered it or we would have figured out a way to you know I would, I would have said: hey here's mine, email me back or something like that and exactly we could have avoided it. But, um okay, okay, I get the point now. Thank you great all right.

A

I think I get the the process, um so we have yeah. We have 10 more minutes. um There's so does anyone. You know um have anything else to add about ee 10. um I I guess um um you know um in enter. Is anyone? Is anyone looking for a task to work on after the current one or anything like that or have any other questions about anything for ee 10, tcks.

A

All right, um okay,.

B

I don't have a question about this: we're going into holidays we're going into holidays. um The final topic here um on this.

A

Let me find it oh yeah yeah.

B

Yeah, um it is to push the the this call into social media. The issue exists, and it needs to be reviewed by you, um scott and maybe others. I think the write-up must be technical um for for it to go live and we should be sending it. You should be reviewed this week. The eclipse foundation takes a vacation after the 17th and the calls for 2022 are 12..

B

The media should be going the day the day of like maybe five hours before, since this call is eight and we have people from euro joining and then it should just be set up for the entire year. We should be tracking teresa retrospective. That did not happen, and ideally my recommendation is aside from the media. It's my condition is: is there appetite to do a retrospective for 2021 in january?

B

Maybe on the first call um use half of it and we can have. We can show you what we're doing for microprofile. We created an excel sheet and it goes by topics you know: do you feel engaged? Did it take too long for a pr to be acknowledged or be removed? Do you feel like all this kind of feedback, that we should improve right?

B

um How is the feeling? Because I believe that without respecters retrospectives on work done, that are ruthless? uh The project cannot cannot improve, and this is for open source awareness and also humble us and around us as we're, not perfect, and we can do better. So, what's the take on that.

A

um I mean in in general, I think that the biggest barrier is um that we need to get past is the refactoring, the tck refactoring, and I think that's going to bring us more contributors more than anything else.

A

um So, but um it it it's, it's um yeah, every everything everything you do in a person you know, do you know doing a um uh you know, explore exploring um you know what we you know you know did did and you know didn't and what we need to do better. Is you know you know it can, can help as well and um that's a good way to find out. You know what you know. What else is a barrier or what other barriers can we you know, can we can we eliminate?

A

Personally, that's what I see as a the biggest barrier is the you know the um guaranteed platform, tcks internals aren't well understood, and um I you know it's yeah that you know solving that barrier with the refactoring.

A

I think will help a lot.

C

A

That's just my you know my opinion on that, but if there's other barriers we should solve them as well. So um I don't. I didn't see that I looked in email. I I looked in the email. I could see an email discussion about this, but I just didn't see in the actual. I didn't see the issue link like the issue number 770.

A

um Let me um oh.

B

Yeah because it's not, I think, it's not crucial and we we plan it for 2022, but clearly we need to have it before the first call and actually it's one message. So if you click on the on the issue, you can see, I added a trust. It just needs to be destroyed and reviewed, and um you know yeah.

A

I I don't yeah, I don't see the get issue link. It's just. It's blue, like it's a link, but it's not a link. It's just. It's not a you know, not a link.

B

Yeah, I'm sorry, I was rushing to do this and I was like I will add it later. So it's kind of I will add it later for you, okay, you can see it and just review it. I I set it up to review, but everyone here can do that, because, if you're joining this course and you want, we want others to join, you know participate. This is a great way to do so.

A

Is it in chat? Is it once so is? Is it in chat, I'm looking in chat now I stopped sharing.

B

Issue is under the tck repo.

A

B

It is the last um git issue opened.

A

Okay, all right! Well, I we're we're near the end. um You know I just wanted to. um You know, wish everyone a happy, holiday and um hope everyone gets a um a break and um yeah some time to spend with um you know you know doing something you know doing other things besides uh computing, um but I also look forward to working with each and every one of you on. You know on tck, so um yeah.

A

Thank you. Scott.

D

Thank you much.

B

Thank you. Everyone, especially you as cats. You, I think one of the best things they have happened to to to the tck this year is that everything goes in the melonist and you have been an example, a formidable example of to emulate on transparency on conversation and prs and issues, and it's fantastic, sometimes no one replies, but we get those emails and I think it's a thankless job and you do it with so much care, and many of us noticed so. Thank you so much and to all of you who choose to contribute.

B

Tck is the most important item, because it's about testing right, so tooling, documentation and without it no other api under jakarta can be compatible. So really hotline.

A

C

A

You thank you very much.

D

All right have a good one.

A

You too, all right: I'm gonna, stop recording.