AsyncAPI Related to AsyncAPI Spec 3.0 release, 19 Jul 2023

Previous Meeting

Next Meeting

⏯

youtube image

►

From YouTube: Spec 3.0 (July 19th 2023)

Description

https://github.com/asyncapi/community/issues/761

A

These things, my myself so.

B

Yeah, hey it's your.

C

B

If it's your mom's in-law at home, it's like yours.

C

A

Exactly except I, don't have my laptop here.

D

A

Have coffee I have water and I have I didn't have to prepare anything perfect.

D

Like in uh all included the hotel yeah, then I don't know how to put it.

A

Late, maybe snack yeah.

C

A

C

Wrist lace, okay,.

A

Maybe I don't know something like that.

C

I don't know I.

A

Don't have it but yeah like virtually I. Do oh look I, have it bracelet.

B

You have one so we are live already.

B

um Let me share in the.

B

In the specification Channel.

B

I'm so lazy that I'm already thinking now, we should have another bot that will post the link in the in the slack. That's.

A

Major laziness.

A

I mean that's developer, laziness.

B

B

Just to clarify like so because you made me a hosting zombie also expect me to be the host of the meeting.

D

A

Soon say so: it's not me actually who's putting fathom in the call okay I can no I can actually I can start since I. Don't think anyone has questions so far. Maybe you do serve here or you cannot remember some things are right because you were on silent all the time so so yeah so.

A

All right so I'm gonna then start myself with something that I have in mind.

A

C

A

A

All right, so you can see my screen right.

C

I'm assuming yes, yes,.

D

Yes, sorry for no, no, your feedback.

A

So yeah uh thing is that I was thinking. How can we better organize the work that we're doing on the specification to the to launch version three right and um yeah? We had this issue that um that.

C

A

Let me try to find it uh actually I think this one with this one I'll be able to find it and as well maybe.

A

And then this one yeah so generous created this issue right, which is well this issue and all of the uses below and even more for each of the templates and so on, so so yeah, some great stuff, great great job. It's a really tough job to do so. I'm thankful that he did it. But um but the thing is that having this this year is good for tracking. Maybe um it's good for us for let's say for ticking the boxes whenever the things are done, but uh it's hard for people to understand like okay.

A

How can how can I help? What's next? What should I be doing? Next? How can I jump in and say uh you guys need help with um uh Studio support, TV3? Okay, where do I start right with this studio, V3 or maybe I, don't care if it's Studio or what else, but how do I actually contribute to uh speed up the release of version three right? So so, how can we make it more visible? So, um instead of splitting by repo as.

C

A

Here right, so we're split in by ripping each of this uh in each of these uh bullet points that say per test.

A

What I'm going to propose is that we go a slightly different approach and we split since it's a very big task. Let's say the whole thing is a very good thing.

A

We take a slightly different approach and, and that um we split let's say we split the whole thing not by repo, but let's say by deliverables by things that can be valuable for the users or at least for the people, uh for instance, for the developers and for for the developers can contributing to the projects and uh and also for those who are working with documentation. For instance, uh why am I saying this?

A

For instance, I'm mentioned in people's explicitly people in writing, documentation for version three, because it already happened that um they're writing specification examples for version three, but they cannot really validate it or it's not easy for them to validate it right. So so, they're here put together a an issue for that and um and I think that that covers, let's say that. That's uh like a a good example of what I'm trying to achieve here.

A

So what I'm going to propose is that um or what I'm proposing already is that we split all these things by, let's say deliverables, so.

C

A

Forget about the number of the cycle here or forget about Cycles. You would uh let's say that I forget about it, has a member or something, but imagine that I call this enabling validation right, early validation of or enabling the validation of uh version. Three right. What do we need to help those who need to start validating version, 3 documents um as.

C

Soon, as possible.

A

What's needed so, first of all, we need to, of course uh it's something that we will need to do is finish. The specification right, which is not yet done.

A

um Cool thing is that the spec itself is done, but since the spec itself requires that distance schema bindings and the parser are also updated, uh we're I'm counting them as not as it's not yet done, because the the parser is still missing for this for two features.

A

For the last two features that we have in the pipeline, which is a multi schema, multi-format schema support and uh trades the new trades mechanism, so everything is done, except the partial support, So yeah, so finishing the specification in this case means just go ahead and finish the parser and add part parts of support for each of them. Once this is done cool, it's done like the spec is done. We don't need to touch anything else in theory right so so yeah and the other thing is that we want to um enable validation right.

A

So, aside from um adding support on the aside from this issues that are already created like this one, that we have to update the distance to my files and so on aside.

C

A

This, let me check uh one of the things that we will have to do is we need to enable validation in studio and in the CLI, because these are the two tools that we're using to to let people's validate version, 3 documents right or at least maybe I- haven't thought if we have other one, but uh but if I wanted to validate something quickly, um Studio or CLI, so help me right and and if we do it, um even if it's in a release candidate on each of these projects, of course, that's not I'm, not meaning that we do these new versions of new major versions or anything like that.

A

We're all release candidates, um I'm already, enabling them to to validate using the CLI or using Studio or web right. So that's already valuable. That's really cool, um so I'm. Putting like let's say we have um again forget about the dates: uh I, don't care about the dates here, I'm just using this tool to to help me visualize the whole thing, but uh but you know we're flexible with this. It doesn't have to be uh up until September, 4 or anything like that right.

A

But let's imagine that is the case that we're targeting September for this right or early beginning of September.

A

um Even though we have holidays in the middle but but yeah what's next, then then it will be like okay, so what's next to to keep developing this asyndbi version 3 support. So the next thing that we will have to do is probably that's. This is yesterday one of the ways displayed it. We could have splited. This in a different way uh is, to probably add support for documentation generation for code generation, for converting to version three right and I want to clarify something.

A

It doesn't have to be sequentially, so we don't have to finish one cycle to start the second again. This is just for us to kind of like visualize it, but of course we can be working on things of the cycles of the cycle too, at the same time for cycle one. But the thing is that if we I think it's important that, if we focus on helping ourselves with the basic stuff which is just validating it like being able to validate it and things in respect which is I, think it's pretty basic, then we.

C

A

Then jump into this and say like okay, we're gonna, migrate, HTML template and markdown template and Generator, so that we can generate uh documentation, and that also means that we have to migrate the react component because it's what the HTML template is using. So if that's the case then so this this only this task will be about um will be about um all these things right um support, converting to version three again.

A

um This is just like me trying to think like okay, what else we can include in this cycle. That is not so much. Maybe it's almost done by the way, so we could already start offering this in studio and um not sure if we have a command for this on CLI. But if that's the case, then on CLI as well um and also in CLI Maybe, uh maybe this could be yeah. We could be figuring out.

A

What's the best way to implement this in the CLI, like maybe with these flags that Lucas you and I discussed the other day um about, like certain Flags, certain commands and say like might be migrated, but some others have not, and we can figure out the best way or maybe this can be just to separate release candidate of the CLI and uh whenever it's done it's done and that's it so so yeah I don't care whatever.

A

Whatever way we we want to approach this so so yeah and then we can keep going and creating new Cycles or new uh um groups of tests, let's say and um but um to come up with. Let's say if, if we just do this so far in these two cycles, this will mean that these two will be done, will not be done completely, but Json schema files will be done completely. This one will be done completely react components will be done completely generated, rsr, HTML, template, markdon template and server. Api will be done partially right.

A

um Because celebrity AI has many more things, many more dependencies, so we have one big task for code generation, but it's a huge big task. We might not want to migrate them all initially right so uh so maybe what we can do here is just to include a bunch of them like I, don't know two or three templates just to show people how you could do it and then our task here our job here, will be more about uh going.

A

After all, the template owners and saying please prioritize migrating it uh migrating I mean so other support for version 3 as soon as possible. Right um and that's it, we don't have to migrate them all. Of course, that's that would be insane um and we don't have to wait until all of them are migrated or supportive versus trade to be able to launch version right, which should probably launch person three with some Java templates, node.js templates and um I.

A

Don't know I mean the code generators, the competition generators, what we need them, but uh I, don't know, maybe go just to say something or typescript or whatever or python, um whatever we decided right, but the problem with server API is also. It also needs the bundler which is yeah. It's the only dependencies. The server API is the only one that reads this for version three to be updated right. So to me, it's a it's a last priority here. Right is at least priority, so so yeah.

A

um So the cool thing is that what I was saying the way, I'm or the reason I'm.

C

A

This is um because, by doing this, we will have all of this done, or almost almost all of this done right except summer API, which will not be complete Studio. Then we can focus on completing studio and enabling full support for with real Studio. This could be another cycle.

A

Maybe or this could be even a cycle of the of Studio itself um and the CLI same thing, so we need to migrate or we need to add support for version 3 on all the commands and figure out a way to uh slowly migrate, this and so on, right, but yeah again.

A

um So that's what I was thinking. So why am I doing this?

A

Why am I doing all these things? Obviously I don't want to do this myself. I, don't want to be migrating all these things myself.

A

I also don't expect that as working full-time um on the postman team right um will have to do all this job ourselves, because it's not our responsibility and it's also not I'd, say- um will not be enabling the community to do their part. So so what I suggesting what I'm the reason I'm doing.

A

This is the Unseen that there's a big mess in understanding what's needed in order to ship or Century, so I'm, trying to put a little bit of order here like trying to categorize, there's the best way possible and using something that will allow us to visualize this in a better way um and also as a way to encourage new contributors and say hey, you want to help with um you want to help with I think eBay, three release. You know what we have this feature here.

A

We we know we have this issue here and we need to add V3 validation support, so maybe on the CLI. So maybe you can pick that you can pick that issue and by the way this is not an issue not yet it's not clickable, because it's not an issue, it's not a real issue, uh but we can create real issues on each of the reports if we want to, but.

C

And we can tell people like hey so just.

A

Pick this one or just pick anything that is here on the left, because it means that it's not yet started right.

A

Nobody has picked it yet and uh and yeah just assign it to you and work on it and instead of having to instead of having to choose from here right, uh because this is cool but yeah like where do I start, because if I come here, for instance, server API and I choose bundler, uh then you're, probably not helping much right, uh I mean you're, helping of course, but it's not super helpful in the early stages of uh releasing version three right. It's much more helpful.

A

If people uh finish this task of support, converting three or finish the parser or you know like, or maybe the react component or things like this, so yeah.

B

um Because I'm.

B

A

B

ah You stopped sharing.

A

Yeah I stopped sharing because I I could not see you.

B

Because I'm you made me super lost because, like so you're talking about deliverables people you show shape it up. Then you talk about prioritization of issues. Then you talk about all these issues, one by one. What are you actually proposing.

A

So what I'm proposing is that we split the things in a different way so right now, all the tasks are split by Reaper right, um but not if we want to release version three and and provide value as soon as possible, not to the end users, but also to other contributors.

A

um Not all of them will have the same priority right because, uh for instance, uh if I want to migrate, if I want to take the react component task and create the the new uh react component, add support for version three in the new react in the react components. Sorry uh one thing that needs to be done first, is that you? Basically we finish the parser, because the parser is not done yet right.

A

uh We need to finish the Json schema files right and we need to release them as always candidate, so that you're able, from the from the react component you're able to use the new parser with free support and then add whatever features you need to add on the react component right.

B

um But like, uh but how shape it up will help like? Can you have to visualize.

A

B

You visualize dependencies as well like so.

A

Visualizing it so the thing is that what we're gonna do here is that on on the shape it up here, so forget about the deadlines and so on. We don't respect that. We don't.

B

Have no no but I mean like how, from shape up I know that react component depends on par searches.

A

So that's the thing so I'm not I'm, not um I'm, I'm, specifically creating tasks that have no dependencies. That's the thing you know. So, if let me share again so I'm trying to split things in a way that or that we have the list dependencies as possible.

A

C

For instance, this one right this.

A

One someone might think like this. uh This is a blocker for version. Three validation support right, but it's not uh so, for instance, adding support on the JavaScript parser, for these features can happen at the same time as this one as we are working on, enabling the support on CLI and students around right. uh There's a dependency here, I'm sorry I'm using my my finger to point to start like you could see, but there's a dependency here saying like hey parser. uh We need to add this to the parser.

A

We need that to add this, to the distance schema right. So.

C

A

Can track if those are done or not right, so there could be dependencies inside this View right inside this only view right, but not between all of them right and um and yeah, for instance, if I um another thing on the cycle: two, for instance,.

A

So, if next, if what we, what we focus on next is to add uh documentation, generation support, then this will be enabling, as on the cycle three, let's say right to uh to come back to studio and say: okay, so we now have the react component migrated and uh we have uh generator migrated some. The documentation generators migrated. So we can now from student. We can already enable uh being able to download uh the documentation only documentation.code, uh because the code will not be done yet um on well. In the case of in this case.

A

Yes, we'll be done, but imagine that we only have that right, so we only have the documentation. So if.

C

A

What I'm trying to do is like we have the we.

C

Have to ship everything instead.

A

Of going repo by repo and figuring out the dependencies, which is a mess to.

C

A

Let's split it in in in a different way: not the report by repo. So let's try to focus on providing value in each deliverable. So whenever so, we're gonna deliver something that.

C

A

Not finished by any means which is Studios studio, for instance. Let me show you, so this is something that circuit did, for instance, so Studio enable validation, rounding right. This is one of the one.

C

A

Disable everything except the editor, Tab and documentation, validation and document validation. Sorry, we should use the parser.js with candidate and this is still available.

A

Of course, the result of this task is not that this is available on studio.ac.com. It will be on a deploy preview URL right, but people can already use Studio there on that display review URL to paste the document there in one.

C

Day right so we're already.

A

Providing value as soon as possible and that will help the dogs folks, who are right now: they're, writing the free documents and giving them to us and and I'm like I, think it looks good I think it's all good, but I'm, not sure if you made a mistake or not so so, yeah like enabling these little things uh will help us.

A

um Let's say like set it clear: a clear roadmap of uh to to get there. You know what I mean.

B

Like so, for example like so when you're showing this issue from Sergio, that's the example of what you want to have right or.

A

Yes, this is an example of what what uh I would love to see that this you see like, for instance, this this issue here, that's actually created on the studio. This could have been created somewhere else. It doesn't mind it doesn't matter involves the species on schema repo, the parcel repo studio and CLI, because it's not per Reaper. It's super value right.

A

So, let's provide a small feature as soon as possible where on studio and the CLI right now, what do we need for providing this small feature on student CLI, the parser and the business schema files as well? We need them cool. So, let's just focus on this right. We got validation, support on students you're, like cool next thing. What else now we can validate? So um what else can I do? I want to enable uh documentation on the studio side by side? What do I need to do? I need I need to migrate.

A

The react component right, I need to make react component support, uh support version, 3. cool, but if we do it, then, let's just not add it to studio right.

A

But um let's, uh let's take the opportunity here that we are migrating the react component and let's make it. uh Let's make the HTML template to use that new version right and the generator as well right, so that we can generate the new HTML template using I mean the old HTML template, but supporting version three and.

C

That's really a.

A

Little bit of more value right because now I can plug the red component on studio, so I can have validation and the documentation side by side. It's already a little bit more right of value. In the same time, I can already generate documentation with generator right.

A

So it's a little bit more value. Then another thing we could be doing is: okay, we are we already now we already have a generator, and some templates migrated like markdown and HTML. For instance.

A

um What's next so next is could be another cool thing that we could, that we could be doing is let's enable people to download code to generate code from studio right, uh but it's only from Studio. No, we migrate the templates right and it's a small chains on Studio to use a new version right and you will be able to download the generated code. But it's a small change, also on the CLI, and you will be able to download uh to generate code from uh from these templates as well, because they are migrated already right.

A

So like this slightly providing more and more value until you get there and they will get like to a point where everything is migrated. Except of going retrograph.

B

Paper right and going back to the shape up, can you show it for a sec? Oh.

B

B

Specification enable V3 validation support. That's these.

A

Are the features.

B

These are Cycles, not features.

A

These are the features that I want to have in this cycle, so this is. We have cycle one here right.

B

It's one-to-one relation or not.

A

No, no they're, independent, okay, they're independent, so I could be working on enabling field free validation, support on CLI on the studio and all the stuff right and at the same time someone can be working on things.

A

The specification, if I finish this one before the other one, which is unlikely but can happen if this happens, then once this is finished, the only thing that I will have to do is to update the parser dependency on the other one, and that's it it's it's a minor thing to do: it's not really a lot so so yeah but yeah, but as invalidation on studio, for instance, on CLI.

A

C

A

Updating the parcel it's uh it's a little bit more right. It's um like, for instance, in studio we'll have to I, will have to understand that this is a V3 document right and because it's V3 in that case I saw block. Let's say all the panels that are not supported yet right. Instead of failing right. Instead of making it fail, I just don't bring the documentation. I, just don't render the outline I just don't render whatever is not yet ready. Right, I don't offer these features, um and then we slowly get started.

A

Removing this uh exceptions. All of this, while in a separate, Branch, I'm, not saying Master uh and releasing this to master right, but we can already deploy this to a deploy preview URL or to a npm package release candidate package for the CLI. So people can use this specific CLI version, and here I'm thinking more about as contributors not end users right.

A

So if I'm writing docs about V3 I would love to be able to validate the documents that I'm writing right without having to pay in France or circuit, or whoever right to validate them visually, and because it's easy to make a mistake right. So so yeah, um you know slowly rolling out things instead of going one by one, because if you go things that the reason I'm proposing this is, if you look at here in each of this Asus, you have dependencies. So CLI has all these dependencies right.

A

We need to make it like complete, full V3 support to the CLI. We need to migrate all of these things. um If we go to Studio same thing, we have all these dependencies right, probably well yeah. Only these dependencies and.

A

Yeah, it's some another another one is missing, so we we are depending on server API as well to download the parent test code. So you have dependencies. This one doesn't have much dependencies. Actually, this one is done. We just need to release it.

A

um This one is probably not having a bunch of dependencies right.

A

um It's only a bag, so yeah rather problem something to fix this one has dependencies. One Parts here in this case.

B

But you know like, um um but what's the problem that it has dependencies.

A

The problem is that it's a dependency help here. We know we know how things are structured because we're working on this full-time, but.

B

If you ask, but you shape it up, you'll also have dependencies.

A

But no I'm removing them. That's the thing. If you.

C

If we're working on.

A

Cycle one, the only dependencies that you will see is here, but it's here, it's clear: it's I have to work on this thing. That's it.

B

A

B

Mean by by this, where do I see that there's a dependency.

A

It's not it's not that there is a dependency in in order to have V3 validation, support right, I have to I, have to implement support schemas on the paraser and then on studio and CLI right. So there's there are dependencies here in between these two and these two CLI and Studio, but not beyond that. It's not going I, don't have a dependency with server API or with the react component or or anything like that right. That's.

B

My point dependency, but what I'm? What I'm missing is because you all the time you talk about like that, you can see dependencies but like how you see it I, don't see how. How do you see that it's clear that they're, depending on each other, this four elements.

A

No, it's here! No, no! No! That's what I mean that's what I'm saying here. It's not clear that they're, depending on each other, but you only have to deal with these four and understand how these four and depend on each other. Only four yeah.

B

But then, let's, let's go further uh down the line and then there's this um react component. That requires.

A

B

And actually parser and um and it's ice I suspect it's in a different cycle. Right, yeah.

A

B

A

It's not as urgent for us right now. That's the thing! Yeah.

B

But then, um okay and and the idea is that, if something is in cycle two and cycle one is not completed, you should not pick it up or.

A

Ideally, this would be in sequence right, so once we find we finish cycle one, we can go to cycle two. You.

B

Can jump component you don't need studio and CLI. That's.

A

B

That's my point, sir. What I was saying.

A

Before is you don't need for react component? You don't need the things all the things that are in the cycle, one you can already start if you want to, of course right, but what I was saying is that you can start in great component. I said it before when I start like you can work on things on the cycle too.

A

If you want to that's fine right before we finish the second one, but you need to be aware that uh then you're gonna find some blockers right like, for instance, the the parser will not be done yet. Probably probably yes, we'll see, depending on the progress on the cycle, one right, uh Json schema files. Well, it's going to be done because it's done already right uh so for direct company, it's going to be done, but if you're targeting something else like generator um well, actually generator will be, will be possible.

C

um If you're targeting.

A

Other stuff, like um you know, like supporting version three or migrating all the templates, like the condition templates code templates and so on. um You can work on this. You probably face some blockers because not all the things are done here on the single one right, but you probably don't face many blockers, so you can parallelize the work.

A

If you want to now the reason I'm suggesting that we split it like this, and we don't take into account, we don't see it fully as a sequential thing, not fully at least right, but let's say like if you, if you are doubting of what to pick next to contribute.

A

I would suggest you that pick something from cycle one instead of cycle two, if you want to right, if you're in the mood, because this ones are going to be more, let's say or are right now more necessary than having a react component right right now, as it is right now right. So let's focus on migrating or or having this as soon as possible. So other folks working.

C

On other tools,.

A

Can at least validate the documents right at the very least right. So that's that's what I'm! That's? Why I'm saying this and then, of course you can work on this, you can work on this. I was even planning on um renaming these cycles and not putting any numbers right. So, let's not put numbers, so um some people don't get confused that they have to be sequential right, even though we have the dates and so on, but again um yeah.

A

It's it's a tool made for shape, but this is not meant for this for, for this kind of uh project management that we're doing here, but I think like forgetting about this about dates. This could still be helpful right in the way that we've visualize things and and also like everyone, can see the progress here of each of the tasks and say like okay, so progress here is zero, so I'm gonna pick this one.

A

So I can help push right and instead of wasting not wasting my time, but instead of being uh looking to contribute to I, don't know the bundler, uh because I think it's cool um and I just pick one from the list, like I'm kind of my I'm gonna work on the bundler, it's like!

A

No, if you, if you don't mind working on anything and you don't know where to start, don't start on the bundler start on any of this, that we're suggesting here like please, let's focus on this first then this then this right, but of course behind the scenes, someone can be working on any of the things of course right. This is just like trying to sit a little bit the direction and organizing a little bit of things. So so there's a clear view of what's missing and what we could be doing right doesn't make sense.

B

Like, like so I I I bought the idea of like uh of grouping by a kind of feature description, um but I mean I still, don't see how it's really resolving the dependency issues still really have to have the knowledge. The.

A

B

It and see like what's.

A

Going on dependency issue is solved. The very moment I mean not solved not fully solved, but, for instance, you have dependencies in this one here right. uh So in order to start with studio and CLI, you need the other two to be done. So there's a dependency right.

B

um So the author defines dependency, that's what you mean.

A

No, no! No! No! No! What.

B

I mean no, that's.

A

Just a fight by luck, it was a chance by chance that that is not it's not related, probably yeah. It was probably in my mind like because or in in Syracuse, mind actually that he wrote first the dependencies and then the actual stuff, but yeah it wasn't on purpose.

A

uh The thing is that, um in order to do a studio in CLI, you need to complete the other two so yeah there are dependencies, they're, not probably not not visible uh here, but probably they're, not visible here right, but it's very limited, while the difference compare to having this issue is you're, seeing the dependencies of all the things that we have to do not just one specific feature but you're.

A

Seeing all that, you think you need to navigate all the dependencies of all the features that need to be implemented in order to fully support version. 3, investing kpi, that's to me the difference. So this is a manageable right, because it's a it's a it's long right. It's a long thing! It's a, and each of them has many more branches like server API, for instance, is hell right, um for instance, to.

C

A

C

A

Server API is a good example. One might think that you need to do all of this in order to be able to ship version three right, um but.

C

That's not true.

A

Like you probably have to migrate to new Js parser, but you probably don't need to my to that support in the bundler or um or converting to version three. You probably don't have to do it initially right, um but you probably have to add support for all these templates. And you know all the templates basically or some of the templates is not all because they are used because server API is used on studio right. That's why and what.

C

I mean is, this is like we know this.

A

We know that server API is in studio right and by the way this is mixing is I have to edit um and that's why I know that I need to work on the server API in order to release version 3 support and make it GA. If you want right, but um we don't need to work on each of the things here. We.

C

Can release version.

A

Three, without all of these things migrated right, so you.

B

Know what I mean so how it will be visible in a shape it up? There would be just one item. One bet forget.

A

Them forget about shape it, so uh what I'm, having shape it there, it's the same as having Asus here somewhere in the community Reaper or in Spec, rep or somewhere I. Don't care in a specific place, not in each trip but in one repo having multiple issues where you have a list of the of the issues that we will have to be working on somehow grouped differently, not by Reaper right, so grouped by feature like we need validation. So there's a group for validation.

A

There is another group for rendering documentation, there's another group for migrating to version three converting to version three, so you have different groups of of tasks right, like value actually actual value that the people could leverage right. We could do this in in GitHub answers. The problem is the reason I'm doing this here.

A

Is that if you look at the at the spec repo there's a bunch of users, if you go to Studio, there's a bunch of issues, so it's going to be those new issues that we're going to create they're gonna, be in a mess of other uses right and they are not really useless per se. It's just like a categorization of other uses. It's like grouping of other pieces right, so uh yeah.

C

They can be.

A

There, but again we need to somehow uh we need to somehow like group them and so on and create issues and so on. We can do it. We.

C

A

The reason I'm suggesting that we go and either.

C

A

From that or or independently, we use this because here you see the progress of each of the items without having to navigate through all of the issues right at a glance you see how everything is is going, I can I can quickly come here and see if something is stuck here right or not, and some things will have been moved forward or not. I can change the to the other group of tests. Let's say it's cycle, but let's not call it cycle. It's another group of tests.

A

Right and I can click here and see if this is moving forward or not or if something is stuck right. I can click here and the same thing I can I can quickly see if there is, if things are moving right, let's move it. It's a visual. In the end, it's a visual tool right, but we can make it work with real issues on, of course, I just didn't. Do it yet because I don't want to pollute the ripples or one repo with lots of issues right. You know what I mean. So that's.

D

Why the alternative to this I would say it's just the GitHub project right.

A

um Exactly the back end to all of these things, it's actually a GitHub project. So all of this is so the shape it doesn't have any any.

A

um Any database oops it.

C

Doesn't have any.

A

Database, so it's all here right, so I created this project here specification right and we can probably be using this- you could be using this as well right same- is that it's a little bit less visual.

A

That's my point right, yeah, um also, something that you quickly see here is comments and events that happened like I, can easily see here. Someone saying hey, I brought I put up a pull request um or I'm stuck with this. Someone can help. Please and I can probably see it quickly, right and and not the same, and this history of of um of messages and events is showing uh all the comments that people are posting on each of these issues.

A

If there are issues, of course, um so it's like seeing all the comments in in all these issues in this new place, because they are somehow related right. Not all the comments on all these issues, only the ones that they want to show out here right. So in the in the comment you purposefully have to go there and say slash progress, blah blah blah right.

A

You need to use the syntax to make it appear here right, so so yeah which is going to change by the way we're gonna make it easier, but um but yeah um it's a it's a good place where you can actually see everything. That's happening in a single place um in a.

D

A

Say, structure or organized way and I think this could help.

D

I want I wanted to say that I'm I I, like um using Cycles like in the standard way so I, don't see why we will jump out of the Cycles in like ordered Cycles, because in in this case, so um I mean what we are trying here or what I see you you are suggesting is to follow like um like uh value.

D

This is called value, driven delivery, whatever.

C

Yeah, whatever, whatever.

D

Scrum Lucas loves, you know. Maybe Lucas knows that.

C

D

Value-Driven delivery.

D

As soon as possible and whatever so I think it's still, it makes sense, at least from my point of view, to keep Cycles ordered just to just to um follow just to act as a clue for people that jumps into the project or wants to collaborate.

D

What is the strategy of delivering this, and what is the most important thing to yeah.

A

To work so I think I think you, you said you say the right word here using it as a clue. Yes using it as a requirement. I, don't think it's useful, so um so having Cycles sequentially It's a Wonderful approach right, it's a wonderful approach, so you will be blocking stuff here in cycle too. That doesn't need to be blocked right. If someone wants to start working on this, why not if they have all the dependencies they need solved like it's the case for uh supporting uh support, converting to version three.

A

uh If the parser is already done uh actually I'm, not sure we're using the parser. Are we oh.

C

A

Are yeah so if the parser is already done, but the cycle one is not yet finished right, then I can already pick this one right. I can only pick converting to version three, so I should not be blocked by the rest of the test on cycle.

C

A

So making it a requirement, no making it a clue like uh I want to. As.

C

D

A

Like this is in our, in our opinion, maintainers of the spec.

D

A

Is in our opinion, this is the most important. That's why it's here. First then, this is the second one most important, but of course anyone can jump and work on anything they want, because they may not even be aware of the whole shaping project here right. They might just go to to GitHub and pick whatever they want and that's completely defined. uh I.

A

Don't think it's going to happen because we've seen almost no contribution towards version three uh other than as uh working full-time um and from the people coming to from coming from the Google Assistant of docs, of course, um but um the thing is that, as a clue, I think it's a good tool as a requirement is really bad right.

A

If I'm, assuming that yeah in theory, we have like we.

C

A

Many engineers and developers working on this as.

C

Much as many as we want right, like the theory, says.

A

That we have an unlimited amount of people working on this right. The practice says that it's not the case. The practice so unlimited means that if, if we had 1 000 uh developers working on this hell, yeah I will not use this. uh This way of working because yeah like let's just identify which are the most the the biggest dependencies, this probably the spec, the parser and the Json schemas and so on, and then the rest can probably be parallelized right or mostly parallelized right.

A

So yeah no problem, but since we don't get a lot of uh contributors right um and then not even not even us, we not, even as we are a lot of that, we are a lot of people right. So since.

C

A

Not a lot of people. What I will suggest here is like, let's, let's focus on something right, instead of being uh completing one part here, another part here, another part here and one month after we still have no value to deliver. uh Let's contribute the parts that are going to deliver some value, even if it's a small one like validating right, but that's already something right and then people can already can already use it and use it for something which is it's already. Okay, it's already validating.

A

It's only validating okay, but it's all in something right.

A

um If I focus my efforts in the react component now, which is what I was going to do before, I started doing this, then I will spend probably two months working on Direct component and during these two months, I will have provided no singles value right and probably I was I was going to be blocked by the parts here, so I would probably have to jump into the parser first, so so yeah and try to to help with the person and so on. So um so my point is: is exactly this like?

A

Let's not, let's not make people Jam from one thing to another. Let's stay focused on something and.

C

Once this is solved.

A

The few people that we are working on this and and actually pushing on this, let's move the focus to something else, and it's not one Reaper or another Reaper, it's it's one value, let's say or deliverable or call it whatever. You want or one feature and then another feature, so we start providing some value right on each iteration right.

A

Another thing that happens here is that uh by coming by not by coming to shape it, but by grouping tasks in a different way in a way that um it's not related to repos but to these deliverables, you could quickly see as a contributor, if I'm looking to contribute to something and I, don't know where to start I will just go to one of the groups and say: oh this one hasn't been picked up yet: okay, I'm gonna, pick it up myself and I know that I'm going to the right place right because that's a group of tests- that's uh probably it's.

C

A

Because the fact that it's uh or a an ordered sequence of Cycles here it gives you the sense of or the feeling of coyote as well right.

C

A

C

A

Most priorities: I can choose it's a little bit less and so on. Right, then it gives you a clue on where to spend your time contributing, so it makes it a little bit easier for people to pick an issue and contribute right and we've seen this on Studio. We haven't finished our first cycle yet on the studio, and this happened on Studio I didn't know where to start with all the things that we put, these things together and I only have had to tell people go to this shape. It project see all the bets.

A

Click on all the the beds that you see on the on the left and so on, and if you see a circle on the left, click on the circle, it will take you to the GitHub issue right and, if nobody has assigned it to them,.

C

And if nobody has requested.

A

Saying I'm gonna work on this then leave a comment say I'm gonna work on this and start working at it and I I only have to explain it once and then people start picking users easily right, but those issues where structured in a way right.

C

A

They're easy to solve: there's they are super small uses to solve right like creating an icon here, creating a button adding something to the design system, whatever something small right that people can pick it up. But if I in my intention to migrate to this new version of studio right, which is a whole project that might take more than a year, probably right or around a year, if.

C

I tell people hey this.

A

Is where we want to go uh this design that I created the figma? Let's try to make this possible.

A

The obvious next question for people is we're doing this. Where do I pick you? What do I pick from here? Actually it happened to me: we got we. uh It happened on Studio that a contributor who was contributing to to some of the uh design system, uh icons and so on. We got we. We basically finished all the all the issues that we had for that cycle, It's Already, Done, Right um and the guy came to me- is like hey, okay, so everything is already emotion right.

A

So there's nothing on the on the left now, so everything has been started. So what else can I do what what else can I can? I pick up and I said in my in my effort to try to explain what.

C

Was happening, it's like okay,.

A

Can you please go to the figma design and if you identify other objects and buttons uh menus whatever? Can you create technician saying I'm gonna create the uh the button X here that I've seen in the design of figma um and I'm gonna? Add it to the to The Descent system?

A

um Can you please come up with this, and um and and do it yourself and I'll tell you.

C

Like yeah, this is really cool. This.

A

Is good go ahead and work on it? It's not perfect, but it's something in the meantime. I find the time to create more issues right um and a week after I had to pin the guy saying like. uh Can you won't help with this? Are you struggling with this, and the guy told me like yeah I'm struggling with this, because I really don't know how to split this design.

A

This big design into small pieces right I have no idea how where to start from, uh and it sounds obvious to me how to do it well, yeah I've been around for while right, so so it's normal, but for people who are who are not so experienced might be hard right and uh and they're willing to help.

A

C

The same thing.

A

Next virtual Studio, which is a big project and next version.

C

Of facing API, which is another big project,.

A

um When you have to split it up in small tasks that people can actually pick turns out, it's hard it's hard to to to to figure out how to split it and which one to pick right. So that's what I'm trying to do here to come up with small things that the people can address and contribute right.

D

And you clarify two things.

C

D

One, how I assign that a task?

D

A

D

Yeah the way to visualize that in shape up like um how can I see from shape up that one task is assigned. Is it possible.

A

um Not yet I plan to add this, but um but something.

C

That we've been.

A

Doing in studio is that you update the progress of that task. Let's say to 10 or something like that and say: hey I'm, taking I think it is at all the one percent and I'm picking this up. So it's not on the zero right. That's.

D

A

Enough to to, for.

B

People to see that's.

A

Why, when I was telling people uh click on the click on the on the circle? It will take you to the um well, not this one, because this one, it's not a real easy yet, but.

D

um And that's the other question I have like foreign.

A

I was telling people clicking the is to check the comments right if.

C

A

Has uh asked that.

D

A

Gonna work on this then just leave a comment say I'm gonna work on this yeah.

D

That's not true, because sometimes you don't put the comment because whatever because you feel alone well.

A

D

C

D

What can you do? Yeah, let's, let's make it official if you we want to assign something, at least for this project, yeah.

A

That's that's what I'm telling people on Studio like if you want to work.

D

On something leave.

A

A comment also because if you leave a comment here and you're, not part of the organization right, you are not part of you're, not a code owner of any repo right. I can only assign it to you here. If you leave a comment.

D

A

Right I cannot assign it to you here like, for instance, uh maybe to you. Yes, because you are in another teams in other teams. But if I want to sign into a series.

D

Yeah you're gonna send to me because I'm on Earth, I guess but.

A

Because you're, a co-owner of this practice and schemas as well so.

C

A

um Yeah but if you were not you probably it's beginning that because you're part of the organization, so probably you can but.

C

A

I, try to, for instance, Samir right. It's not there.

C

Yeah yeah because I have.

A

I have them added to one of the projects, so I forgot about that as a as a for with triage in permissions on Studio. So yeah, that's why that's that's! That's why it's uh so he has three engine permissions to so to be able to rearrange issues on the projects and so on, so so yeah.

A

um So that's the thing! um If you leave a comment, then your name will.

C

Appear here and.

A

I can assign it to you on GitHub.

D

C

Other question.

D

C

Other question.

D

Related to this that I think it's important for people, especially for for owners, um is, should we create an issue for all of those like the the current screen that you are showing.

A

Yeah I think we should I think I think yeah. We definitely assured.

D

I also believe, because you can yeah, if you are the product individually, it will shown properly here. Yeah.

A

We need to show it because if we don't show it here, if we don't create this as an issue, then nobody can leave a comment. There. Yeah.

D

Because there's.

A

No issue right so then you will not have intermediate steps. You will only have not done done. Yeah.

C

A

And no and everybody will have to leave comments um here on that main issue which could become a mess, because the.

D

A

D

uh What are you talking about? Yeah.

C

I I I I thought you were yeah.

A

I'm not doing this on purpose side, because I wanted to validate this with. All of you, like uh I, think this could be a good idea, but.

A

I think that uh if we, if we end up choosing this way of working, um this is this should become Asus, definitely and even finish specification issue. But it's all becomes right. um In my.

C

A

Oh you, even though this one is not so important, because the actual work happens on this tasks, not on the group itself right, so so yeah like here like there shouldn't be so much communication here right uh the things the communications will happen on each of this right. If they were easy, of course, so yeah you can probably have some add some updates here about something but yeah like.

C

A

Necessary, it's not necessary because shape it is going to go through all of this and it's gonna. It's gonna update the progress here and it's gonna show all the comments here right. So so yeah you don't have to work on the on the parent uh issue. You don't really have to comment there. If you don't want to so yeah um so yeah. Definitely they should become issues. If we want to maximize the the communication, let's say.

A

Thank you, I don't see Lucas, but because we'll be probably like.

A

I was I, wasn't looking I wasn't seeing your face, Lucas and I.

D

Was like what is living the whole conversation I can tell you yeah, probably.

A

I know that he doesn't like uh this is still.

D

A

C

Yeah I, don't like what.

A

This kind of project management, stuff.

B

I I talked to you at the beginning, I like when. If it's there's gonna be one link to visualize, I I, like it doesn't matter, doesn't matter what the tool is I'm, just questioning, because I don't think it's solving the dependencies. It's still you're still gonna have to click and go through and navigate. So it's definitely not.

A

The advantage it's it's, not it's partially solving the dependency issues, that's the thing. So, if.

C

A

Look at the whole project. You have.

C

D

A

Is a small part of it? Then there will be dependencies, but many like way less dependencies, because it's just this part. You know what I mean so right now in this common in this issue that DNS created using all the dependencies, you have to click through all of them right.

C

So what we're going.

A

To do in on in.

C

Using these approach is that we're gonna we.

A

That we know all the dependencies and how everything works, we're gonna cook it so.

C

It has the less dependencies.

A

As possible, so whenever someone is going to work on something they, uh they will face this problem, the less as possible right, um so they face the less dependencies as possible and they will not have to like, for instance, like I, said, there's a dependency between uh server, API and bundler, but.

C

A

Gonna work on an invalidation support on CLI and On Studio that dependence is not relevant to me because I'm not working on that on that. Yet you know so the dependency is still there, but.

C

I, don't care about it yet not.

A

Yet I will care in the future once we get to address the survey, if you guys stop right, but in the meantime, once while we are addressing validating a file, um I, don't care about those dependencies. I don't need to see them. Yet that's my point! So that's how it's solving it. It's hiding something dependencies because they're not important. Now you know what I mean. So we're done we're going steps instead of trying to tackle everything at once.

B

um Is there that can I the view of the progress and completion? Is it there? Is it also available per bet or I have to.

A

B

So, on the left side, you had bets on the right hand: side you have. The visualization in Cycles do I have a view where I can actually see, for example, without Cycles everything. The whole.

C

B

Because I, basically don't care about the uh the the dates and the cycles and.

C

B

A view where I can click on one bet and also don't have to go through all the Cycles.

A

No there's nothing like that, because I mean the tool is intended to be used with Shape Up approach, I mean Shape Up, uh you don't have to Cycles. At the same time, you only have one cycle at a time right. The next cycle is not yet decided. So that's why, uh but, given that that's, why it's not I, don't know if you hear my my daughter's screaming inside big.

B

C

A

Like that, but um what we can see that it's going to be merged, probably today or tomorrow, um is the progress on the left side. You could see all the bets right so you'll be able to see the progress on each of the bets without having to click on each of them to see the progress right, so you will be able to quickly see on a given cycle on cycle. One I see all the the progress of each of them right without having to click anywhere, but only the ones in this cycle.

A

If you want to see the ones on the next cycle, then you have to go to the next cycle right. So well, that's why? Because in shape up you don't have a you, probably have the next cycle, but um the view is different, very different. Let me try to see if I can show you an example of how it looks like with another I'll share the screen as soon as I find it.

A

So studio, for instance, Infinity, would be a good example, although I'm not sure we have another cycle yet yeah, it's great okay, so I'm gonna share my screen.

A

It's funny at the iPad warrants you like we're. Gonna share the whole screen three two one. It's like come on, you're just crazy, so this is the case of studio, for instance right. So if you look at the next cycle, this is the view right. You don't have progress, because these are peaches right. These are not beds, so we are pitching ideas for the next cycle right now and at some point they will be selected or rejected right. So we will decide like okay, so this is going to the next cycle.

A

If another one appears here this one, not another. One appears here is that yeah this one? Yes, so this is the view so far, because this cycle hasn't started yet right and.

C

Because it hasn't.

A

Started yet there's no progress because you shouldn't be working on it because it probably hasn't that's but that's shape of stuff, and the problem with this is like I know that we're not doing shape up but the tool you know the tool might be still useful for for many of the of the um for many of the things that we need to solve right, even if, even if we're not using shape so so yeah. That's that's it! That's that's why you don't.

A

You cannot see it and that's why it doesn't make sense to see like the progress of each of the bets for all the cycles, because.

C

At the very moment, the.

A

Cycle is finished it's. This is either dropped or move to the next cycle right. This is either not drop, because this is either complete and therefore it's not interestingly interesting anymore uh or it's.

C

A

To the next cycle, because it's not uh it's not finished in real shape up, if you don't finish something you drop it, but we're not gonna do this so so yeah foreign.

C

B

Once the bet is completed, it will disappear from here.

A

No, it will just so what we're working on adding it's already. We have a pull request so today, or tomorrow will emerge, so there will be an indicator here of the progress of the completion right here in this card, so you will see a line like 80 percent or whatever right, and then you will see that uh the line is here until the end and it's green. Now it means it's completed right.

A

um Something like that. So you will see instantly if something is not completed like this. One, for instance, will show a full line.

A

Probably green right, I haven't seen the design yet so I cannot say um this one will not all right. This probably will be I, don't know somehow red, because it has two items before the 50 percent so and it's already certainly late right, probably so yeah um this one will be completed and showing the full line in.

C

A

And yeah and this one will be right: the red line here with 20 or, in this case zero one pixel right, something like that. So yeah.

A

But if you want to see what's in the next uh like, what's the completion state of the next cycle, then you just go to the next cycle, not in the studio. Obviously, because not the case but um sync oops.

C

A

I'm not sure if it's 21 or 22 yeah.

A

C

A

Since we have the future Cycles already enabled.

A

You could see the progress here instantly right and if something, if something is done or not, we can even I'll be happy to show the progress here, maybe like, uh because this one is 100. This one is 50 and this one is seven percent, then 100, plus 100, plus 50, divided by three. The total progress is X could be there as well, but per cycle.

A

We actually I, didn't tell you, but uh we added support for not having dates at all, so the the dates disappear and the Cycles are just group of the tests.

A

I tried to do it like this, um but I think we're gonna, remove this from from the same bit too, because it's really making it a mess to maintain is it was just really for us, like you, shouldn't, be using this tool if you're not doing shape up right, probably we're using it as a because we have nothing better right now, but.

C

A

A

That's not the aim of the tool.

A

By the way, we're getting contributors and shape it.

D

I just admit that I need an issue for a few minutes ago.

A

Oh yeah, yeah and just.

D

Complainer you.

A

Know and the official you're, the complainer, the complainer.

D

But I made my first PR the other day, so thank.

A

God yeah, so oh, the link is wrong. It's pointed in the mind to my repo, oh with you, thank you. So this is.

A

A

So yeah we got the pull requests. This is what I was talking about: the bad status, completion right.

D

The bad thing here is that you will not see the the play preview because of the authentication with GitHub.

A

Yeah yeah, that's something that we have to figure out as well, because of the authentication of the GitHub, app and so on, and the display preview robot URL, but yeah I have an idea on how to make it work. But that's another topic, but this is respectfully zero call.

A

C

A

You have ideas on how to improve it for our case as well. Let me know, because yeah, it's open source- it's controlled by by yes, so.

D

Yeah, this is actually why I submitted an issue, because I mean right. You were showing this I noticed that navigating through cycles, for example, was like a pain because if they are not sequentially, you know like how can I go to the four or whatever, without having to click one by one yeah.

C

Yeah yeah, but.

D

Yeah so yeah, but globally, I I, like the idea to we have something in place. It's gonna improve, um especially for people that comes, and it's like helping um contributors to start contributing potential contributors. So I like the idea and I'm gonna I'm happy to to implement it.

A

D

We can give it a try what.

A

I'm trying to propose here is that we have nothing to lose, so, let's give it a try. If.

C

It doesn't work for us.

A

We drop it, that's it, but um so there's no compromise here so just a I think this can improve the way, the visibility that we have and um the visibility and and not visibility for. Let's.

C

Say, like visibility for the manager.

A

Courses like that invisibility.

C

A

Even for people who want to contribute like okay, this one is not yet done. I'm gonna pick it up uh among among a bunch of them that uh are available. I'm gonna pick this one because yeah it looks like a priority. So.

C

A

Makes it easier for these people as well, and um so it's visibility in that case right? That's what I mean and um and I think it's positive, because when you are creating a bed in this case a bad issue on the left, you really need to think about. How can we provide some value with the next step? Instead.

C

Of just let's just migrate.

A

The tool- okay but yeah once.

C

We finish everything cool, yeah, yeah, sure.

A

But we can start providing value before we finish anything right. So that's my point. Yeah. Even we can release version 3 without having to finish everything right like I I. Don't think we have to finish. Let's say maths, code generator or some or I, don't want to say that net is not important, but uh we probably don't need to have all the code. Generators migrated in order to release version three right, I think we should have a bunch of them, so people can already try it out. You know.

C

A

Java in Python- something like that, but that's it not all of them, not all of the Java, maybe maybe some one of java, one of node.js.

A

C

That's my point like.

A

Completion is not the same thing as release date right, so we could be releasing version three. Even before we complete everything right fully right. We can and this my my making these bets or groups of tasks, it's forcing us to think about how how we want to group things together to deliver value as soon as possible and.

C

A

That's positive: either it's sacred or it's a databases or whatever I, don't care. But it's forcing here.

B

Okay, 30 minutes past yeah.

A

I said I was going to be fast, but.

C

B

uh Will you later pink with the recording folks from the docsfox, because you're not tomorrow, right on the call.

A

D

Yeah I'm not gonna, be able to join, um but all I wanted to say is that um I found several things always like same issues in all the documentations that they are. It seems it's not very clear that they are working with B3 or so like they provide most of the examples based on version 2.

D

and the documentation mostly is based on mission 2. So I will should I write this in the in the meeting agenda and you will discuss even if I'm not joining or.

B

Or just write in the in the docs Channel because they will probably like the first thing I already have. My, in my mind, is which issues you mean exactly.

D

I wanted to do that, but I didn't want to to some like I. Don't know how to I, don't know how to write it without sounding, like not aggressive.

A

D

Yes again, you know it's like.

B

That you see the issue in some. Unless you see it's just one out author always so.

C

Just no, no, no! No! No! No.

B

A

Had yeah come on, yes, just just buy it on the on the onions, Channel, don't wait for calls or things like that. It's not neither. Sometimes it's just a message. Just you know, sort of.

D

The ones that are writing sorry and ones that are writing documentation for V3.

B

D

A

Think so, yeah sorry.

B

In in Doc's Channel, they are older, yeah.

D

Yeah I know I'm asking who, who, if it's only Google summer of talks, people who write writing.

B

D

B

There are also a few folks from uh the mentorship.

D

Okay, yeah because I have a bit of me. Sometimes: okay, yeah.

B

Alejandra is coordinating that.

C

A

B

Bye, bye, yeah.

C

A

C

A

You have to freeze.

C