 Welcome to the September 26th areas cloud agent python maintainers meeting lots to talk about. There's an agenda. We're recording Any trust policy is in effect for the Linux foundation and the hyperlegic code of conduct is in effect Let's get into it any announcements adjustments things people want to talk about We can now send nonverbal feedback. Yes slow down from reactions on the toolbar Are you glad? Really zoom let us know that. Let's see where to do it. But anyway, all right Am I editing? I am not. Sorry with that. Okay PR 2486 is needed For an upgrade that at least we have to do I don't know others Need it basically we have a very old issuer that is in production and this upgrade fixes some tags and Other issues related to issuer revered records. Basically, we've had a problem with Every time we do a revoke. We've had to fix the revocation Meaning we have to sync up. What's in the wallet and what's in the ledger and that comes back to this once we go to 010-2 We wouldn't be able to fix it. I believe or something like that. So anyway, we have to get this fixed in We'd like to do a 010-3 with just this fix So I wondered a is there anything else that ought to go into it and be any objections to it and see Shanjot being sure that we can actually do this as a separate release. Is that doable as a 010-3 Shanjot, but that'll work Yes, there's no dependency with the code that's there in me right now. It's all like I think it should work like no no new commits that that have a dependency The status on this is basically from yesterday the unit test that was pretty like in the fixed ledger entry function That's what I have fixed in this PR. It wasn't unit test covered earlier So it has covered now and the peculiar logic with the escrow profile session That was proving a bit tricky to do it in test has been resolved So that test has I've added that test gen just squashing the commits into a single commit just like Minimizing the get commit history. That's what I'm doing right now and should be good to go in next oven Okay Comments from anyone on us doing a 010-3 The only comment I have is that we have Some changes that we'd like to see in a release pretty soon as well I don't think those are quite ready and if this is all ready to go I have no issues with supporting getting out at a 010-3 release But I definitely probably would like to continue pushing on a 011 release after that one comes out Okay. Yeah, and that was the next which is to look at this So 010-3 with just that one right now, you don't have any others that you need in there Daniel So the big one that I'm pursuing right now is the the mediation routing keys fixes and that is going to come with some changes in the behavior in out of band I think at least Which seems like it would probably not be a Patch release I suppose so yeah I don't know why my fancy Why can I not edit it? Why is it not there anyway? I did a nice fancy Here's all the releases since then and got it all set up, but it's not showing up. Okay And even that's not working Let's look at merged Man, I'm not having a good keyboard day Okay, the merged I came up with one that would eliminate the ones that were put into the they're only in Maine and And So Basically, we'd be talking the ones but basically everything on this page Move that code did exchange version set. So none of these these are just sort of maintenance things The peer did resolution. That's a big deal bug fix documentation documentation and their container Obviously big thing we're getting in here would be just SD jot. So that's a significant enhancement That's clean up update pie did is clean up. So Doesn't look like 11 would be that big a deal. So that's good It's got a lot of things but not that mostly to do with developer improvements Yeah, the main thing that we're pushing on right now Is that the routing keys PR but that's pushing us in the direction of Compatibility between our out of band and did exchange implementation with At least recent releases of AFJ. Yeah So right now we don't really have the ability to do a did exchange protocol with AFJ because it's using did pure one We don't intend to implement did pure one, of course, but they're they're working on a did pure two and then This also influences our compatibility there because we also don't have the ability to do public did out of band With AFJ because our routing keys are inappropriately Formatted on on the ledger. So did you go on the ledger? Yeah Because they're unqualified. Yep. Yeah, and then we've got this you saw this, right? Yeah, right. So I Think the right out of sync Yeah So the RFCs I think were probably a little confused themselves. I Don't think it was clear to the authors at the time what the difference between the did key reference and the did key itself was Yeah, I think the intent is clear that it should be did key references, but the examples are incorrect So what AFJ is doing is accepting either a did key or a did key reference But it's currently sending out Just did keys but only an out of band in the did exchange. It's it's using the fully The full reference did key and routing keys Yeah, so I think we can probably do something similar in that where we're willing to accept the just the did key Yeah, but then we'll just transform it to the reference so we can process it Yeah, as if it were a correctly formatted value and then we put in the same fix in AFJ Right, and then we're good to go with switching over to using references. Yep. Okay Coordinated updates love them. Yeah Okay And you're dealing with this right now Yeah, I've got some tests that need to to be updated. I think this is the main failures there right now Okay, so this is that's the main thing you want Um Well, so then the next thing that we really want to push on after this is getting the did peer support in For did peer to and did exchange So those are those are kind of the two big ones Okay So that's that one This one that one I'm less Invested in having that one pushed in quickly, but it's just getting rid of some old stuff that was unused anyways So it's just a cleanup. Okay Shantja you've got this one in work in ready for review This is Underly going testing right now. So I got the initial review But after that I had to make some changes because of the issue I discovered So yeah, yeah, let's be let's keep on the convert to draft one necessary This one This one is ready for you. Okay, and then this last one Are we gonna close this one? Are we gonna move forward with it? This is really old. So Should we just close this one? Or is this a useful fix I Haven't looked at the code recently, but I would have to guess that it's probably hopelessly stale at this point Yeah, and the effort to resolve merge conflicts might be Yeah, yeah Shantja, is it a good idea to close this one? Do you even remember it? I don't have to like take a look again, but it's like whatever at that stage I started working It was stopped at a decent stage. So I'd like it'll be Like if we need to we can just recreate early reopen this but for now, I think we can close this All right Is there an issue that is linked to this one to track the the actual problem I believe so Oh that one resolve no error message if tail server is down. Okay. It was there Yeah Thanks Yeah Yeah, I guess I keep these open at least it sounds like it could be like a very helpful for some Agreed it can be very helpful Yeah Okay, that leaves that one more issue In the review required which is this one Right. This one is one that I've had on my to-do list to to look at deeper Kind of the main challenge here is that I think there's a better way to do this I just haven't had the time to think through what the better way would look like and to suggest alternatives here So, all right, it's fine to leave it. I think Yeah, let's just leave it so our review requires our Let's get a look at those. Let's keep using draft when things are not ready And we'll figure out what to do We'll and then we'll plan a 011 release Just as a matter of interest where are we how close does that get us At some point we need to go over what's left in the You know in this Set of things for a IP to I think we've got that one nailed Whoops Daniel, do you know way more about root coordination than I do? Do you remember where we are with that I? Believe from a protocol perspective, I think we're actually In a good position on the coordinate mediation protocol now I think the only thing that was out of spec was the did key usage We are Cutting over to a state where we can accept either base 58 or or did key We already were accepting both of those actually but we're storing them as base 58 hence the issues with being pushed to the ledger in the incorrect format But I think from a protocol perspective. We were already. Yeah. Good to go there used a key that was The update About this the encryption envelope. No Okay, I have not looked at this recently. No, yeah So on the did exchange one So we've merged in like the the did resolution for a did peer to and the storage of did peer 3 as aliases to the did peer 2 But we're not yet able to use those in our did exchange Protocol so we've got one extra step to complete there before we're good to go on that one and Sire tuck you are not going to be available Jason For at least how do you know how long you're still going to be on MDS? after you come back so so Jason's off to Utica New York to referee a international across event which is pretty wild And then but you're also doing MDS we can't hear you if you're speaking we can't hear anything you're saying Jason if you are at least I can't I assume others can't Something wrong with his things put a chat I missed that so Hopefully we can hear from Jason in a second here But if he's unavailable to work on that and I get through the routing key stuff here Hopefully pretty shortly. I can put some time in on Getting the did peer to And did exchange put together. That's a high-knife priority for us that I think they they want me to push on that so Yep Um, I didn't realize Andrew you are on the call Would you be able to take a look at where we are with 587 and sort of you know, don't do it but sort of give us a Here's the state of it and here's what would need to be done. So spend an hour or two on that to Give us guidance us to where we are Yeah Most of the functionalities there in in Oscar. Yeah One of the sticking points is just going to be doing resolution during did resolution during unpacking which I would probably restrict to Methods that don't actually make any external calls personally You know resolving did peer and did key right? Oh, hold on. So that would be so you're saying You would not use For not use Public goods Hard to say and I'm sure there's a use case for that too Um Yeah, I'm I'm aware of at least one scenario that my Interested parties would have they would want to see did web be able to be used in a did-com perspective. Yeah, I know So going back in my memory a long way, I know Um That's something that matter to bias always used to say they wanted to be able to do is use the same did for all the did There did calm Yeah, I mean, I mean it's implementable. It's just like You got to hold that hold on to this message in the queue while you do the resolution and If you get a hundred of those at once Yeah, it could be a problem. Yeah Anyway, um, I think you've got cashing though, right? We do yeah, we've got a cashing layer in there. So the resolution would be subject to cashing It would okay well, so assessment of What was needed and where we are with that? Yeah, you know, these two were both dropped as I don't believe they were ever Merge, right? This was just closed and yeah, both of these were closed So they never were applied. So the question is if they weren't applied, where are we with them now? That's funny to remove a mitt. Yeah, we just did that I think Well, we're about to we haven't done it yet But we will and and those two would be just we're not gonna change the we're just gonna change the flags to default to being on We're gonna keep the flags and default them on give no way to not use them Basically the change. Okay So that one certainly take this is likely taken care of this one was a larger change Remember how many? How many updates? Yeah, there was 20 file changes. So there were a fair number of changes to do with these I don't know how many were Yeah, so some of them were real versus Or sorry, we're code changes versus test changes. So Anyway, you could do a bit of an assessment of where we are From an AIP zero if we are that close zero eleven might be one zero zero So yeah, I guess we're we're not looking at doing protocol support for did come to yet, but No, but we're getting close to being able to With the With this the rotate protocol that's in place Still a draft but That would enable us to get ready To transition to did come to okay, cool Okay. Yeah, if you could give us just a written update of what you think we're at that would be I think you're the by far the best to know at the envelope later where we are Okay, thank you. Can you hear me now? No, I can't Yeah, right with the integration work I'm working on just the backhand stuff. I'm hoping probably within a week or two of work I should be able to have that stuff polish and ready for a production. However Again, I'm off a couple days and then there's probably some stick handling And if they want me to do some front-end if they want me to build front-end buttons and stuff for it So I can see it stretching out all the way till maybe end of October Which would be which would then allow me to dive back into the did peer stuff. Okay, that's a long way Yeah, if they if they want me to build in the UI components have them like be able to update and error correct and things like that But I don't know exactly whether They're gonna need me to do all that obviously the less the better, but I also know that We want this built and if we want it built more than they want to build it then Yeah, okay, I'm probably the one doing it. So yeah It'll be probably a little bit more time before I jump back into the peer which is Unfortunate of someone to see that cost the finish line Obviously, I can't wait and if Bloom's able to take it then he should cool Yeah, I think I should be able to Put some time in it well before the end of October. So I think I might be able to leapfrog you out there. So, okay Okay, we've looked at the PRs If you don't can we extend this meeting a little bit more Does anyone mind I mean, I sort of assumed we would I think we need to change this to an hour long meeting anyway and Cut it short when we can I'm gonna have to leave in two minutes, but I I think you guys can carry on without me. I'm just listening not very helpful You're always helpful. I mean, I know Um any Okay, to keep going. Yeah. Yeah Okay Awesome that this is closed Daniel. Thank you for that great work. I'm figuring out exactly what was happening here very clever So this was submitted by Ariel from From the AFJ team Two services Acropy wasn't resolving them but the reason was because these IDs are identical and so when we were Serializing we were dropping one of them because they were the same Nice work. I noticed he's already got a PR into AFJ to fix their side of it Nice Okay, so 020 dev1 of an on credits RS is there which eliminates which uses the latest CL signatures So this is now ready available and so should be merged into an on credits RS Which last I checked was oh also on Daniel shoulders not good So at some point this needs to this will need to go in I suspect The peer did pieces probably higher priority, but Daniel up to you to figure out what what works for you I think either I cut out or you cut out for a moment there, but I think I got the gist of what you're saying um Yeah, the Did peer stuff is probably gonna be my my first priority for a little bit at least But I'll see if I can squeeze in some time anywhere for getting that merge taking care, okay? I'm pretty sure I talked the whole way through that so it wasn't on my side Bob shared components version We now have again tagged releases for ask our We're automatically picking up indy VDR and indy credits in aqua pie Or sorry indy credits are automatically picked up so indy VDR and ask our need to be added So those dependencies will be updated Imagine Uh shanjot that probably would might be one once you get through the last of your Work on the other side to to get this as a transition task We'll see how that goes Um I can I can probably put in a graph PR just to test Okay, those uh because there's a dev release of ask our Yeah, up that we can test against And it would be right that was the other part of that discussion um Should we go I I don't know how soon Zero three zero ask our will be out should we actually do a release where we rely on dev one or should we wait until I would wait. I don't think it'll be long but um One thing I'd like us to Validate is the postgres 15 Yes, okay on kubernetes because I can only test it locally. Yeah okay Okay, so Then andrew if you could do this upgrade with a work in progress and flag it for way he can run the Testing on the upgrade. Okay. Awesome. Okay. Good. Thank you And then this last one Yeah, this might be another good one to put into an upgrade again I don't think it affects us. This was a conversation and basically they're flagging that we're using a a dependency that's Got a vulnerability reported and it would be good if we could get past that um So it would be good to get this upgrade done sooner than later Uh, is a am I right in that any any objections? um And I'll just flag it as a high priority. So we've got the the button on it Um, I haven't seen if they've made any breaking changes Yeah, that's the obviously that's the whole question. I I'm not even sure what version I guess we're on Yeah, anyway Obviously somebody would need to run this and then see how we do with the the integration tests Typical dependency updates those are the ones I flagged any other ones people have to fly from the issues Did we look at 2501? I noticed that you had it in the list. I don't I don't know that I saw you pull it up Uh, oh I didn't pull it out. Um, so this one is just a Fun dependency dance. Um, so we recently added the multi-format librarian And there's an issue with using typing extensions greater than we're equal to 4.6. I think um The issue has been open for a while The workaround isn't too painful. Um, but it's Starting to cause a bit of a headache with resolving some dependencies. So Yeah, I think that typing extensions there's another one as well that There's a like one version that matches up between all the other dependencies and stuff. So yeah Yeah the The other thing that's kind of exacerbating this is we've also recently pulled in the peer did python library from sygba Uh, which also is perhaps a little overly picky of which versions of things. It's expecting Um, I've submitted a pr to them. Uh I haven't gotten much of a response from sygba yet. Uh, I'm meeting with them tomorrow So I'll I'll see if I can knock some heads and get things moving there Um, yeah, there should also be a pr for me from like a month and a half ago or two months And I never got a response which is Jet having the library generated peer three because it doesn't even know that dude peer Thieves a thing and it's a pretty simple algorithm So I tried to add it as a method to to that library and got no response So, um, yeah, they haven't been super responsive and if you want to poke them on that one too Yeah, I think probably the right answer is to encourage them to move it out of their repos. Um, yes Just not sure where you're exactly to encourage them to move things I guess. Yeah Are you a member of their organization? I am not. Um, well, I am in the github sense, but like I have very little I don't know. That's right. They make you a maintainer on that. They probably could And I could I could probably push for that as a stop gap. Um, yeah, uh I'll talk with them and see what's possible and uh, see if we can get things moving there so we can Uh unravel some of these dependency conflicts. Yeah, okay I should probably open an issue actually, um explicitly stating that While we've done the pited update on the zero ten two branch, we haven't yet done it on main Because of these dependency conflicts. So I'll open one up just to track that that needs to take place still Yeah, I've lost one of my topics in here Yeah um Yeah, let's hit this one first Um, so we've had an offer from disney Which is not disney disney Which is I think both the company well, it's a it's a linux foundation Uh foundation disney foundation Um, and I got a note from somebody involved in that that they want to implement open id for vcs in Um In acopai and they're looking for guidance on on doing that they've done a evidently They've made a few prs to just minor things, but they usually they basically use acopai as a black box And just use it So They are now interested in adding open id for vcs and they wonder Are wondering what guidance we have for them in what to do um I've invited them next week to I invited them here as well, but I figured it was too early anyway So I'm not bothered that they're not here. They said they couldn't make it, but um They're likely to be able to make the acopai next week um Thoughts on I believe we have talked about making it a plug-in. I have asked them for what admin api they They would like to have um, so what what interface from their controller they would expect um so Basically, we have the option of figuring out. Okay. Here's how we suggest you do it um, I I know Daniel you talked previously about doing it as a plug-in. Is that still where you're leaning? anyone else have other thoughts on On what to do Daniel So we've got a proof of concept that we've been working on for open id for vcs. Um It's The way we've structured it is we've actually had it be a separate service outside of acopai and we're just making use of Like the jwt sign and verify endpoints Yeah provided by the admin api to perform the crypto um Take advantage of of you know, secure storage all that stuff. Yeah And then just implementing the open id for vci protocol within a separate context We're like We're still pretty early stages. I would say with that I don't think we have a problem with being more open about our our progress and work on that so far but we we have been kind of Behind closed doors a little bit so far on that And most of that's not because there's anything I actually secret there just Yeah, we're just, you know, don't want to air the dirty laundry out. I guess So, uh, yeah, we're we're still working on that. Um, we're to a point where we're interacting with uh, the spherion wallet Which has open id for vci support Um, just because it was there and available and it was easy enough to pick up and and use it to interact with things Yeah So that might that that I think that would be something that we'd be happy to demo on like an akapod call Soon ish I'm not sure if next week is too early or not. Um, I'll have to talk to adam and char and see How they feel about that. I guess not to put adam on the spot here Um Well, I think for me to do connect you guys directly with Sergio Um, who's the person that's been talking to me directly Um, I certainly wouldn't mind talking with them and and sharing what we've learned so far at least. Yeah. Yeah. Um I can Probably join in as well. I've done a quick proof of concept Open id Issue in some presentation also with the spherion wallet. I think it mostly only supports the Jwt vc, what is it? Jwt. Yeah, jwt vc drop profile at the moment Yeah, it sounds right Or the sometimes it's called the jff drop profile. I think So here are the disney Foundation members Um, so I I think I've heard of this. I think it was mentioned at the linux foundation meeting notice. It is a linux foundation Uh community if you will And info search certainly is Well known in our community Um, not too sure. I'm pretty sure I've heard of nimlab before Possibly gaya, um, but it's a it's a decent possibility. You might be able to get, um You know, you might be able to be used, uh, you might get development help on this so That could be a good reason to do it So out of curiosity since hansher mentioned The proof of concept. I'm curious in in your approach. Have you been doing it? Similar to how we've done it just outside of akapai or or what is your approach been? Uh, I wasn't using akapai okay, so it's uh Just a it's like one one python file that generates a a jot and sends it over Cool, um and has the oauth endpoints right the paradigm wallet ought to be an option as well I was running into some some strange issues there that Will probably be fixed soon, but Um, I was able to issue a a jot to it yes We have plans to uh to use the afj open id for vci client package that they've got and try to set up some automated testing on our side um and Test that out as well. Uh, haven't gotten to that quite yet though And I assume this is akapai as an issuer and verifier great Yeah, yeah In the very literal sense where it's just signing the credentials and and uh performing the key look at the look up and uh signature verification. Yeah. Yeah Yeah, I mean we can and probably ought to track things internally as a protocol with states the way we do Um over didcom But uh that would Require some api changes if it's done externally so we've This is something that I could I could see going either way to be honest. Um, so far we've been um Pursuing it and like so it's a companion service to akapai and we use akapai for the issue of verify steps but then also We're expecting a controller um to the The service that we've put together and just like how akapai delegates the business logic to its controller uh similar model where we're delegating like The determination of exactly what credentials are on offer um to the controller and uh Letting it handle the storage of those values as well So regardless of how the you know the values intended to populate the credential are are being obtained whether it's through Like a form submission on a website Or if it was pulled from a back end or something like that, uh, we've been delegating The reporting of the actual values to ultimately be issued um to a separate controller service Okay, I think you can also You can do it mostly statelessly by Using like an encrypted jot in the id token I think that's right. Uh, so you You know you can put all the the issuance parameters in there And then they will be submitted by the client In the final step Right Interesting. It's just the state parameter. I think it's called I'd like to take a closer look at that Um the I guess the the bigger barrier there is if you want a small qr code Then you probably want to shorten that to just a reference right um So in the flows that you've been implementing it sounds like you've been doing the pre-auth flow as opposed to the authorization code flow Yes Right because those are also the only ones that are implemented by saryon and the other way. Okay. Yeah. Okay. That makes sense um Yeah, so I guess to summarize the topic a little bit here. I think There's there's obvious tie-ins to akapi and there's ways akapi can support doing open id for vci But i'm not sure that I would have that be a direct responsibility of akapi still Um, and we've had it be slightly more decoupled than having it be a plugin So far. So it's just something that depends on the admin api yeah Yeah, I mean if you can create jots and sd jots through the api, I guess you you can just glue it all together at the controller level And we can We can add more of the glue later on for akapi to do those things automatically Yeah, for sure Okay, um, what I'll suggest is how about this i'll i'll create a dm conversation with the three of us and Sergio And then we'll tentatively say at akapi next week. We'll talk about this Um, but base it on how you know where he is and what what value you think you can get from him Does that sound right? Okay, and then if if we don't think you know, he's actually got Got much then we'll we can defer it if we feel like it cool okay, um Sherman, uh, do you want to give us a quick update on where you are in rs I saw lots of closing things yesterday and merging of pr. So I was happy to see that What's next? Um I think i'm gonna try and put the rotate api in Oh, I thought it was done. Sorry No That no because we did that after we started the non penciling. So that's going to come in Hopefully that's not Any kind of trickery or anything? Um Yeah, hopefully it's pretty straightforward. Um And then there's still like all the pie tests have been commented out Which is a chunk of work that I'd like to get into in place to Remove ones that aren't needed and get the other ones updated So that they aren't just forgotten about Then there's a big one that I think should probably be tackled soon and the reason I bring it up is so it's Quite diverged from me So that's I think 24 11. I think it's the ticket number For that So that needs to be done And then that's also the work that I think Andrew and Daniel have to coordinate because there's going to be that chunk of work for The that Andrew did with the stuff So I'm suggesting we bring that I'm going to be away In two weeks so um So one punting it off on other people Is good for me, but also that will be a few days where I'm not doing any work Yeah, oh, I think 506 that I believe that's 2506 which is upgrading to an on credits and Sorry upgrading to an on credits to zero two zero and in doing that um Adjusting the revocation end points to deal with the fact that tails files are no longer needed for issuing and And revoking but some additional Fields are so the the things that Andrew did when when credits was upgraded on the mains on the main branch Yeah, so I guess there's a few tickets. I'm just like there'll be on The 10th through the 13th. I won't really be working So it's a good opportunity to kind of get that branch cleaned up without me trying to you know, I think it'll be tricky enough getting it Kind of in sync with me and touching up all those missing things But it'll be easier if I'm also not actively working on something so Just planning forward if if um, we could get the resources together from the 10th october 10th through 13th to tackle that I think would be ideal so there's not two different things going on Um, I think that also happens to be IAW. It is. Oh So it's a bad. It's actually a bad time Okay, well, that's too bad I just thought it'd be ideal that I wouldn't be like bashing away in there and causing conflicts in that. Um, so maybe we'll have to think of that rethink that but um Other than that, yeah, I There's a lot of little things like say that the cleaning up those tests and stuff and getting those into sync um probably We've kind of skipped over the endorser side of things Um, yeah, yeah That is the only I think that's the last big thing left Yeah, and then then it moves into how do we migrate there's a lot of tickets about You know migration processes and things like that that have to be squared away, but um, yeah, I think The endorser thing is going to be a fair chunk of work. Um, yeah So right now there's a bunch of all all the Yeah, there's a whole bunch of B2E tests that are commented out because they rely on the endorser and um It would be nice to deal with that. Um Yeah, I mean things are going they're going slowly It's one of the big problems in the revocation arena is that I have I have no idea what other than the concept of revocation I have no idea what the contents of the tails like I have no idea what any of that at a technical level means So it's it's uh, it's difficult. It was difficult for me to know what was going on and what I was ruining or whatever, so Um I could have given you a full overview of it Yeah, it's just I don't know the code level, but I definitely know all of the concepts and what's in it and how it works I should probably get a still get a brain dump of all that just so that I know going forward what what At a technical level kind of what's what's happening. Um So that was a bit of a that was a bit of a slower thing than I think it should have been but that's Maybe my fault for not saying hey, I don't I'd like to know what This means that means but um, so by moving on. Yeah, uh, the get the rotate one in there um, get the tests kind of in sync with whatever is going on and then, um Yeah, the endorser is probably the next big thing after that. Um, but if we can't coordinate um The merge for main and the that cleanup while I'm away. Um, that's kind of huge as well So we'll have to figure that one out in short order Okay Okay All right, I think that's everything And we're pretty much in time anyway anything else we should raise at this point. Um One last thing that comes to mind So we've talked about creating a plug-ins repo. We haven't done that quite yet, right? No, um We do have a developer. We're thinking of putting on that as soon as he becomes available He's getting his first taste of aqua pie um the repository but um We weren't able to shift him into that already but the idea was for him to create the repository and begin to Foculate it with existing plugins Okay, uh, just thinking thinking ahead, I guess a little bit to a one dot o release Perhaps it's a little too ambitious But it would be kind of nice to be able to drop some of the less critical things into plugins things that are, you know Use case specific for instance mediation Yeah, uh, having that all be contained within a plugin would be nice, uh Also, I think it would be nice to be able to move the connections protocol out Um as well. Yeah Yeah, I had that repo that was like a proof of concept of what we wanted out of there, but it's been kind of sitting around waiting for Such an opportunity as to actually populate it was something useful. So Yeah, it'll be interesting to get that in I know there's there's a couple of small really small plugins in traction that aren't really Traction specific so like change the connection alias really small things basic message storage Stuff like that. That's not really like say traction things and just kind of little enhancements that They they can probably move out of there as well Um, just so there's few more real simple examples for people that want to do stuff and contribute. I think but yeah I like your idea of starting to move stuff out um where possible and uh Yeah, that would that be be interesting and also good because I think we can That way we can also tighten up the core acupuncture code as well. Yeah, exactly. Yeah so Is the goal for this repo to house all the code of all the plugins or just like references or installation like Yes core What we would call core plugins official plugins official plugins things we don't want in may um in acupuncture, but we do want readily accessible and and such that people can add additional plugins that Make sense I just I just very easily predicted in like two years will be like this repo is too big We've got to split split them all into their own repos um But if if it's if it's yeah, if it's being um narrowed at least um, that makes a little more sense. So, okay Just wanted to yeah, I mean when I heard that My other side fear is that we get a thousand repos and they're all different. Um And and that too becomes unmaintainable. Yeah, no, that's fair. That's fair No, I think I think it's uh, again And also like we can draw the line and seem like oh, this shouldn't even be a plugin at all Like let's just make a native behavior, right? Like exactly or this is too obscure that you can create your own repo in some place else Yeah, no, I think I think that makes sense. Just wanted to raise that. Yeah I almost want to be able to use it as like the graveyard for for deprecated protocols as well and And in that case, I would want to see things like be removed altogether eventually But until we're to a point where you know, the whole community has migrated away thoroughly enough Then it's available at least as a plugin and we can put breaking change Into the changelog and say here's exactly how you retain this if you need to Yeah, you can restore it by by including the obsolete plugin. Yeah, that's a it's a good framework to transition away from the core functionality Yeah Awesome. Okay. We're out of time Thanks all It was a good meeting. I think I will extend this to an hour and Um For our team at least probably cancel Suggest we don't have a Tuesday stand up and For our part we'll do this and others can figure out what they want to do We'll see Anyway, I appreciate everyone being here And Happy delightful rest of your day. We have some things to do Thanks