Tools Community Call 16 - near/dev/hub

Captions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.

1. 0:04

   so hi friends my name is ludu and I'm de contributor at near devhub I would like I would like to welcome you to the to community call those are informal sessions where we share updates discuss and present tooling projects from the ecosystem if you want to get involved join and reach out in our telegram group today we have a special special session that we're going to discuss wallet selector near API Js among our the

2. 0:31

   regular updates and today I am very happy to be joined by um a lot of great names around our ecosystem including Mike Josh Vlad uh Bowen who is head of protocol and ne1 gilmo as the head of devil trentin and many others also we can see down in the corner beh below is a heat nice to see you so um the discussion topics as I said today we will focus around the near API GS and wallet selector uh which um current like wallet selector currently had uh well recently had

3. 1:16

   another update and we see that um they are um starts to become a little bit cumbersome not only for the people who are using it but also for the maintainers and speaking of which we have uh trentin on the call and I would like to uh bring you up trentin and if you can tell us a little bit more about of what in what state is currently the wallet selector yeah uh thanks for having me um we're we're really excited about talking about this especially to get some input

4. 1:49

   from the community as well um there's several telegram groups that you could do that we'll talk about that here in a moment um so we just did a release uh it's a moderate release 810.02 just went out here yesterday um you can uh check on GitHub there yep there the you'll see the release the release has a copy a link of all the PRS that went out in it um we have a couple new packages um and um it's uh the the current state of is is

5. 2:20

   that it it's still working really well um it it you can loaded up uh there's the um react there's a couple different test um harnesses basically built into the into the repo where you can just kind of run a local react server a no a node server I mean with like a little react app that like lets you test out the Walt selector so um there's instructions in the readme on how to do that it's actually really straightforward um and so it gets tested

6. 2:44

   um my friend GMO uh was really instrumental and helping us do this last release and doing a lot of uh testing on his local machine and making sure that things were flying um we got some new packages from meteor a new react um parser um so there's just some stuff uh some you know improvements um we're shooting for a new major release a 9.0 release here very soon um that's going

7. 3:09

   to include some breaking changes so um we're going to be kind of uh merging some things into the dev branch and uh asking our partners in crime out in the world to uh pull that down and test it with their their packages and kind of see what's uh what might break and what we can do to stop it from breaking before we release into the while but this next release is going to be is

8. 3:32

   going to be really important to to test really well so I'm doing a lot of communications there's a few telegram groups out there um you can start by going into the wallet Builders Group in telegram um that's a good place to to ping me you'll see me posting there pretty frequently um I have built a private wallet selector Development Group in telegram it's invite only I'm trying to keep it the spam down um I don't want a lot of advertising in there I want a lot of shout outs for new Co offerings and things like that so it's a

9. 4:01

   it's a private group just for devs if you are a Walt selector Dev or you'd like to be one or like to integrate that package that's a great place for us all to kind of crowdsource info uh challenges integration problems that kind of stuff uh it's so that's a good place to do that um so yeah the near the near tools community group is also another place I'm I'm I'm shouting out now in there

10. 4:24

    for new releases of the Walt sctor to keep everybody on the same page as well um I also do announcements on Reddit we we do have a near protocol Reddit I don't know if it was started by anybody in the near protocol it's kind of the Wild West um but I do shout out in there uh once when I'm doing releases as well just to catch new people and see what they're see what they're up to um but you know Walt selector is a pretty good State one of the things that I wanted to

11. 4:47

    point out is that we are looking for a new group of maintainers um so I'm I'm a lone uh Cowboy here trying to keep this thing rocking and rolling um I've got some Help from My Friends I I've I've joined near Dev so I'm a senior TPM at near Dev um and so I've got some new partners in crime that are Deval experts in their devs and and so they kind of get it so it's it's it's kind of help

12. 5:10

    helping me out people reviewing PRS and making sure that we're not breaking stuff so that's been great um but we really would like a like a team to be to be working on um and uh my RFP I'm G try to find a link for it for you here real quick um give me a second the I'm going to post Link in the chat for this um I was I wrote the RFP with with help from people in the infrastructure committee and from from um devell from devhub um and it is it's

13. 5:48

    really particular so uh it's it's created a little bit of of controversy because um I'm I'm being really picky about uh what I'm what I'm looking for for someone to help me out and part of that just has to do with the fact that just like any code repository where you bring in a bunch of opinionated devs and they start plowing away at it over the last couple years it's been maintained by different people so your co-base starts to get a little bloated there's some

14. 6:19

    technical debt there um and so I would really like to see um some Shoring up of of this the codebase increasing test coverage is important to me um so that people when they're integrating a new package I'd like to see some real good integration tests so when you integrate a new package you can just run the Su of test and get a really good idea of whether it's going to work or not um I

15. 6:42

    would also like to see Improvement in the in the continuous integration continuous Development Department um right now it's a very manual process um we we use a pretty typical U main branch Dev Branch development process um but then you have to go in and manually bump all versions the packages right so is kind of a mono repo right where it has the core package so uh wallet selector core and then you

16. 7:06

    have a whole series about 20 25 packages that you can uh Implement on your version of Walt selector depending on what kind of wallets that you want to support with your with your D so um each one of those packages separately maintained but and and then added to the to the wallet selector so each of those versions have to be bumps obviously there's interdependencies there so those dependencies have to be looked at um and

17. 7:31

    I'd like to automate that a little bit more so that um you could use like GitHub actions and published mpm um and and kind of uh have some automation of there running the tests making sure everything's squared away um so there's there's definitely some lwh hanging fruit here for someone to take this on and make a proposal um the link that I just shared in the chat where you can read the propos uh read the request for

18. 7:54

    proposal right so this is kind of a spec and then uh submit a proposal and uh you get a grant from the infrastructure committee to uh take over the maintenance of wallt selector um and one of the things that we're hoping to do is taking this input from the community so both from end users but also really importantly from devs right from people that are implementing this package that want to see more functionality want to see

19. 8:18

    compatibility um we would like to see that uh see that improved so we're looking for that we have stuff coming down the pipeline uh like you know multi-chain support and near intense um that we're going to want to see some integration with as well so yeah but uh it's I won't let it go so this is like this is my cause it's something that's very important to me near and dear to my heart so um I'm I'm always keeping my finger on it even while I'm busy with

20. 8:48

    other stuff so we we'll keep putting releases out uh hopefully weekly or bi-weekly uh we're still integrate packages and we're going to uh look for somebody who is at least somewhat passionate about it that will take it over and and improve it so it will improve in the future yeah amazing thanks a lot um one of the biggest issues that we had for a while was uh the dependency mismatch between your APS and wallet selector and

21. 9:19

    that's been like a very big beta if I might use the correct term for it uh so I'm I believe that the entire Community is very happy that now this is out of the way what are the the immediate problems that you see for a wallet selector at the moment well be it with near near APS or like solely by wallet selector package um yeah I think we we fixed the near API adj dependency um I I would say it's kind of a tossup between the um

22. 9:58

    automating the build process um so that in the publish process um but like I said I I think that one of the things that be really helpful um would be like a a solid integration Suite so when somebody puts a new package in to the waltet selector they can basically spin that up run the tests and make sure that their integration is going to work um with with with everything so um for instance we just um we fixed up The Ledger integration for wallet selector

23. 10:30

    right so now you can just use the wallet selector connect directly with your Ledger um and sign a transaction which is really fantastic because I I I'm a ledger user um I don't know if I'll address you but so so uh it's nice to it's nice to have the integration um so yeah I think I think that's the lwh hanging fruit I would love I the other thing that's I think is gonna be really important is just a review of the code

24. 10:54

    um from both a security perspective um and from a code quality perspective right that's why I'm really excited about finding a team that can dive in there because I'd really like to see some refactoring going on just to make it to make sure that it's it's it's up to speed with our standards in the ecosystem and our security standards so um yeah it's uh it uh uh no Vlad I don't I don't think it does I Ledger doesn't and and Ledger

25. 11:25

    The Ledger core the the The Ledger wallet Vlad asked if it supports named accounts and it doesn't and neither does the internal Ledger live right The Ledger Live app itself does not support near named accounts and we've been um talking with both the group that helped us integrate Ledger uh into the near ecosystem as well as The Ledger folk themselves but it just hasn't been a high priority for them so we're still

26. 11:51

    poking and asking for that I mean there is no need for support from The Ledger it's s because you're talking to the ler device basically um the yeah it's it's a matter of implementation inside the wallet select our UI but let's take it offline cool amazing thank you uh since uh as you said earlier in the call we are joined by some of like the very active Builders uh around near uh I would like

27. 12:24

    to ask anyone who has uh like something to add or they had some Fe feedb back for wallet selector U just a reminder this codes also be shared within the wider community so uh I think if you have any ideas suggestions or opinions I think it's good to have them out and so more people start to collectively think into the pro hopefully the right direction so if anyone wants to uh want to add anything for wallet

28. 12:55

    selector at the moment that would be about n the time otherwise I think we have a lot of opinions here that's very good so it's good to have opinions of course uh like I imagine that the the truth or like it would be something in the middle but um having all of you on the same place and like sharing them all in the same call I think it's a nice way forward so the hit yeah so is it uh uh the key management and kind of the like if you want

29. 13:35

    to kind of move from one wallet to another wallet those those mechanism is it part of wallet selector because traditionally it's very very complicated so like a uh like it's a nightmare so but I'm I'm wondering like a why that should not be part of wallet selector because is it different for different wallet or yeah I can I'm a huge fan of like having packages and utilities and libraries that generally like do one thing and one

30. 14:08

    thing only and the thing about the wallet selector is that it it's literally like helping you to select a wallet right like there are like eight different wallets 10 different wallets 15 different wallets they all have their own interfaces and Etc and to be honest already dealing with you know like the problem of like being a package that allows you to right like you just import something and now suddenly you can

31. 14:32

    choose between 15 different wallets uh actually it's small like right like we have the we are cing with r 30 something right like that's already like like an important problem to be solved and handling Keys is more like on the wallet itself already right uh or maybe even for like another application where it's like hey Lo in here with your wallet we help you to transfer uh but I I I will personally yeah keep it away from the wallet

32. 15:02

    selector because so it so the wallet selector does not need to do three different things right it does one which is like allow you to connect using whatever wallet you want to your account uh Trend I think that you raise your head up there yeah and as the person that project managed the transfer wizard um uh which was a mechanism on the Walt site which was supposed to allow you to do that transfer uh we also had a a key

33. 15:28

    cleanup me ISM but it was largely confusing and caused I in my estimation probably caused more problems than it helped so um as as it stands right now in my opinion um it's a lot easier to take your full access key or your Ledger and go find the Walt that you like in the ecosystem we have a bunch of them um and and and move your account over and and I think by being deliberate about it

34. 15:52

    it also is is is more secure as far as from a user perspective there's a lot less user area involved there um so get to choose the wall they want to and just import your key it's it's pretty straightforward in my opinion so um yeah that that mechanism is very difficult to facilitate and uh I have a lot of experience trying to facilitate that and it was it was it was it was rough

35. 16:14

    so I wonder what what this move from wallet to wallet even means because the wallet does not Ow like own the account and uh if if you have the access key if you have the uh seat phrase or private key or Ledger you just go to another web app and you log in inut the account there so I'm a bit confused about this um um communication I think we need to edicate users more on this aspect that they can actually go to any UI out there

36. 16:51

    I'm I'm wondering how like I think uh emails is a good example and at the same time I want wonder if it is obvious for people that they can use different email clients or they just stuck with for example Gmail mobile app and that's uh all they think they can use uh so maybe the same problem maybe it's different um I have other opinions but I see Sasha wanted to comment on this specific

37. 17:23

    topics so stop here Sasha yeah thank you nice to see you guys uh one time uh I I have like my my my question is kind of more more organizational one right the I've been with with near for more than four years already and for me personally it's always has been a little bit of a struggle right to to provide my feedback to the to the to the entity which accepts it right like this is like the story of any early developing project but

38. 18:04

    anyways like like for now like we we we have that document like it's it's a little bit like weird but still you know like this is the purpose right to initiate like people just throw the ideas like maybe they major pain points right uh to the document without polishing it like without creating FP whatever like all that thing like so uh just to to to get the honest uh feedback from people From The Trenches and my question is like for as as this state like with this wet selector do we have a

39. 18:37

    product owner like or like no right like yeah yeah and not at the moment and that's what we are actually looking for yeah yeah yeah yeah yeah exactly no no don't get me wrong right like I'm like in the same the same p with you guys right it's not just like like some Sasha random Sasha came fromia so like the uh okay so yeah so so what what is the next step like like do we wait for the for the for the person that somehow like appears like

40. 19:10

    one way or another and and Lead this uh this direction how I see it uh is that Tren already prepared the RFP with a bunch of his uh experience and details about how to manage it right going forward and his recommendations published it as part of the infrastructure committee RFP and uh we need the team to we need to find the team uh but obviously probably there is no such a team that is immediately available so we need to grow

41. 19:47

    the team out there and the ideal way is to bring so this today's call is our next step after the RFP is to discuss the challenges that we all foresee for anyone to take such a big project on their shoulders including mentioning that it's a lot of product work uh it's a lot of uh Discovery work it's a lot of uh uh probably user research and user feedback collection from teams like so

42. 20:21

    there are users of wallet selector right and there are new project that are emerging out there that are not using wet selector so it's obvious that there is something missing and we need to find what are the features that are missing out there for example uh near ihub and near intents they're not using the uh VA selector uh at least not directly and uh yeah uh today call was uh the idea behind it was the idea to collect the

43. 20:56

    feedback uh capture this uh recording and attach it to the RFP so it's a good reference for anyone taking the notes and uh What uh yeah we're saving them a little bit of time on doing this research I hope that's answer that answers the question yeah yeah yeah almost uh so like like uh if imagine we are in the perfect world right like like uh so can you can we ask for back train

44. 21:30

    know anyone else so I don't know like crowd crowd Source the initial step so like somebody can uh go through the document know you know like and create like at least like a very simple you know like one like like we can vote for the for the main uh painful stuff so should that should be addressed like immediately or yesterday or something like that like like so we can move forward you know like not just you know

45. 22:01

    like have another Co another doc another telegram group and and I mean is it's a request for proposals so that kind of proposal is what we will will be waiting on so yeah up to the and I can answer I can from from your perspective it would be it would be very it would be amazing if for instance I could set aside some of the the priorities I have dive into the Walt sector co-base and come up with like you know

46. 22:35

    my wishlist you know plan I just I mean I I don't currently have the bandwidth to dive dive into that to from you know for full-time unfortunately um but sure um it would it would you're right we need a plan so that's why we're trying to find a group who will dive in absorb the c-base understand it and then do that product work right right so get get a better idea of like what's the low hanging fruit um and what you know what's the highest priority and then prioritizing those things and and then

47. 23:08

    coming back to us with the plan and then when I be come back to us I mean come back to the community as well right and say this is what we discovered this is what we want to do going forward what do you guys think and getting feedback into what the priority should be I would like to hear two two sides like code code base and and real world right like the the features so let's let's coordinate both if possible yeah thank

48. 23:40

    you yeah um since we have uh vadir who is active Builder on near we also have Elliot and quite a few other people I would like to ask for other opinions or like other issues that active Builders have been hitting regarding wallet selector and then uh we can talk a little bit about near API JS as as well because um unfortunately in the recent months it has seen less less L uh than it needs to and with the current developments uh it

49. 24:17

    would also require quite a little quite a lot of work as well and we actually have some solo effort like Mike's who at one point try to reinvent it himself in order to to work a little bit to work to make it work better for him um but more on this a little bit later so anyone else from the Community Builders who would like to uh address stuff around wallet S no that that Al just one thing is like recently for

50. 25:03

    projects um there isn't really an interest in selecting multiple wallets for at least different products being built so like that's just something to consider for Consumer applications do we actually see us pointing them to wallets or yeah creating you know we we kind of like offboard we offload a a lot of the onboarding onto specific wallets and I think that kind of gets a little difficult especially considering n's account model I think that we can

51. 25:38

    like I don't know I think we could think better about it maybe just a little bit or and then um the other thing was Mike happy that Mike's in here because dude man SP all this time you I I think the uh you know I was following what you were working on uh and you really brought to attention that that whole types how we have pretty terrible type definitions across all of NE API JS and wallet selector where

52. 26:10

    they're not even like real type definitions they kind of just like conflict with each other or their classes something like that um it's it's tough working with near API JS and wallet selector because of how none of these types fit together don't really or you're importing a package just so that you can use a type or something like that um it's it's really messy in that way so that might be a pretty easy First grab is if we could

53. 26:42

    pull all that work that Mike did that ultimately I I I don't know uh where it has ended up but I think that that would be great to use and pull out all those TP Happ to demo later yeah right I I do want to say just really quickly kind of in response um like this is the way I'm thinking about this because I'm like deep deep deep deep into like travel strip land now my my brain is like um a lot of the stuff that happened with wallet selector near API JS is like

54. 27:17

    very weird so like we all learned it the hard way and there's no expectation that we should have thought that this would happen right so like you know I won't go into too too but like the class system like when you're passing around class instances even for typescript types like JavaScript like really screws you over um in a really weird way and so I like I could talk about this for a long time

55. 27:42

    but like I I'm just mainly imparting that like it's not like we messed up like this stuff is hard making like a plug-in system is hard and we did it like kind of the wrong way with classes and that's okay but then when you do it the wrong way and you find out then you just move on you know to the new one so like there's no blame going on this is like super duper hard um I just wanted to say that because um I don't know I was so excited to see Elliott and I wanted to clear that

56. 28:08

    up yeah to recognize the fact that it's hard and there are a lot of uh different approaches and uh the state of wall selector is uh very influenced by the state of nears that is why um whenever I'm talking about world selector I'm usually also bring up the topic of nears and how it plays a crucial role to uh make things easier or harder if it's done not right or right so and I I'd

57. 28:40

    like to just say that there are like uh pure technical challenges uh like adding more features to the world selector such as signing messages uh met transaction support or getting even getting the signed transaction back from the wallet and uh then be able to um send it over as part of the application on the application Level instead of wallet doing it for you and then you have more control in about displaying uh the

58. 29:15

    actual status or um you know pretending that it's already went through and uh do optimistic uh uh like application uh display of the information going forward uh but also there there are challenges on the um like a product side or coordination site as G already mentioned there are over 30 wallets out there and it requires uh each of the wallet to to contribute uh something of those support they have different platforms uh different use cases and

59. 29:54

    different set of features supported so it's a nightmare and also there is a need for decision from someone who will maintain the wallet selector how to enable as many wallets available out there uh for the developers without asking the developer manually picking one at a time and testing and retesting each of the wallets and ideally even upgraded um smoothly transparently without needing the the the developer to upgrade their uh code and redeploy um

60. 30:31

    but maybe that's too risky if we upgrade those uh outside in if frame or something maybe that would break things so there is a lot of challenges on the way and as Mike pointed out it's not that we just messed up it just uh it is really hard problem to solve in the first place yeah one of the community feedbacks that we have seen in the public channel was like a suggestion is to basically well not drop but not

61. 31:02

    endorse uh wallet providers who are not maintaining up to standard uh like requirements for example I don't know be signing transactions or supporting my transaction so what not so I think that uh this would help of putting structure of at least what we are providing for the users to experience but also like as you mentioned most of these wallets have their their specific use cases so this

62. 31:35

    would be a hard one to even this like although that I see it as a um Direction forward would be hard thing to implement but um yeah kind of worried about that I feel like the wallets are like waiting for you know like the maintainers to like fix it and stuff and if they heard right now that like we actually need you to you know like it's sort of the other way around right now like I think it's pretty obvious that things need to get fixed so I'd be really careful about like telling the devs who were here when

63. 32:07

    they could be at a much bigger blockchain that they're doing something wrong you know we can't do that in my opinion it's like our responsibility and it has been our responsibility you know so we just got to like do it and it's going to suck or it did suck or you know part of it sucked but I mean solution would be to introduce better like API level uh for like for mobile wallets we do this for web wallets we do that and

64. 32:37

    then each wallet uh would be just a configuration of the end point and that's it and everything else should be basically uh Universal and then we can drop dedicated support for each wallet out there and replace it with just link in setad of Icon and the link I do want to point out too that like so you know we had a fast near like a whatever Retreat here in Portland and like got a lot done and um um God what was I GNA say about this

65. 33:17

    what were you just saying for sorry H all right I'll come back to this my in might talk um there's a lot to go over yeah well um we agree that uh we I well we all agree that there has been like these issues between uh near API JS and the wallet selector and we also I believe that we can agree that if we are thinking about it as a stack near apig is the underlying architecture if we can

66. 33:49

    say that uh needs to be addressed and type system is uh um consistent type system is one of the issues that we seen we've seen like previous pushes to I think per like I might be wrong but was it near JS which was kind of a spinoff of uh or at least like a different version of um near APS if I'm correct so I think that this brought up like uh some different perspective of how things might be done which would

67. 34:20

    lead me to your effort Mike uh if you can talk a little bit about what you have been doing on site I've seen pretty lent Twitter posts but also like on the to community call where you presented your work but if you uh can show us what you have been experimented experimenting with I think that would be nice awesome um yeah this is actually like really weird actually so like two days ago is when they finally was they like came together and

68. 34:52

    it's like this almost killed this is probably like maybe the hardest thing I've ever done like I B off way way more than I could chew way more um but like I somehow survived and I have to take a nap after this and I was up till like 2: am. trying to like get this ready but it's um we'll still do the the ceremony here so like it's kind it's almost like hard to explain like I didn't really realize how this worked before um but

69. 35:16

    like I'm sure we've seen so on the browser or the console on the left we've probably seen like oops buffer. from and you try to do that in your um in your console right and it's like I don't know what the hell buffer is right and like I've always kind of like sort of known what that means but it's basically like there's like JavaScript that's like ecmascript and that's like the module type that's like you know nodejs kind of um stuff and then there's

70. 35:44

    like this the browser environment so there's like the browser environment and there's like the node environment and the brow and the node environment has buffer so a lot of these cryptographic um like web3 libraries and stuff are using buffer and like you know they're doing a bunch of bite stuff and so they they using node.js and um actually it like last year at proximity I was trying to do something and it was so hard for me to I I just could not figure out on near API JS and

71. 36:09

    so I uh did a bunch of stuff to where it's compiling down to commonjs and it's compiling into an iif which uh you know we can talk about later um and by the way shout out to guo I I see that you're doing some similar stuff with uh the borch which really helped by the way um so all all that to say like here is like this is like really really strange but um because the entire library is

72. 36:34

    compiled down to Common JS this sucker can now find it so there is no buffer so it's never going to throw this crap right so near. viw we're gonna we're gonna call Barry club inside the freaking console and like this is like the actual code like this is what you paste in your JS is this so I didn't even really realize this was this was like G happen um but you can do all kinds of stuff like that so then this is the login I'm

73. 37:03

    just I'm just making it create session key um so Berry Club so we're going to create session key and oh this little thing popped up um so this is like a wallet adapter and then there's a wallet adapter widget package as part of this monor repo and um so this is an iframe and actually Vlad and Eugene know way more about like if frames and stuff like the whole wallet adapter selector thing

74. 37:27

    is too much for me to think about right now so but so it's working I don't know how but it's in the code and so you like click on my near wallet um going to sign in with mike. near and then we're back at this little like demo and so even here so we're going to literally sign a transaction no let's just like paste that see what happens pending update included in the blockchain executed boom so it's actually signing everything like inside there which is really really weird and this is exclosed

75. 38:05

    this is exposed by This Global near so you can just like do near. viw um you know and I can call a different smart contract if I want to it's just like oops I should probably do 08 so you can just do 08 right in the console anyway maybe I'll just like pause for a second because this is like pretty weird anyone have thoughts or like seen seeing this before I love this uh boss like

76. 38:29

    apis at the same time I hate them uh I I can explain the this this Simplicity hides a lot of uh complexity and uh it's great when you are dealing with this initially but then whenever you want to have more granular control you will be on your own that's usually that's where near pgs seems to be have started uh with like let's let's do magical and simple and we ended up with the current state of things um

77. 39:02

    maybe I'm run and oh yeah I mean I'm glad you mentioned that it's almost like a softball I wish I had these Co Snips in front of me but like so you can freaking um I actually want to move this somewhere but you can go like near.

78. 39:13

    reexport do borch cool now you have freaking borch cool you want to call borch serialize there's borch all right awesome you want to use a borch schema cool you near. yous. schema boom there's the Bor schema now sign transaction cool you want the sign transaction borch schema boom it's right there so you can like encode decode create a transaction sign it inspect it you can do all that

79. 39:36

    stuff um it's not like there's like oh you have to use these methods and we're doing a bunch of secret stuff behind the scenes in fact I think I'm exporting too many functions um I'm just like every you have access to everything um because I think that's the way to start right now um especially just like kind of as we I feel like what this is is just like a safe point you know it's like really weird and hard to say like ah we don't know what to do with new APS and it's like we just need like a save point like

80. 39:59

    restart you know can we do it like this way and so um the API though I think uh to your point for all I think it'd be really cool to have like even like little shortcuts so it' be like near. V is like vew near. VT is like vew testnet like don't worry about the config just go to testnet dude like little things like that so those are those are small like things that we can we can talk

81. 40:22

    about but um yeah anyway uh this is this is ath uh that is sort of finally broke ground and I'm going to publish uh this today and I guess maybe this is an okay time to you know like uh you guys want to do it AB nice that's good everyone close your eyes okay cool I I'll I'll I'll pause that peopleit can talk so is this build on top of near apjs or is like completely written from the ground so I was planning on taking mere

82. 41:08

    API GS which is the version that I like forked basically and then we got together and Eugene like I don't know he just has the stuff in in his head and like like we just wrote rewrote a ton of it um but then there are some parts where it was like um especially the wallet adapter widget where I'm using mere J s Imports instead of near JS because those can't compile down the way

83. 41:31

    that we want um so I would say it's maybe more like 60% rewrite than yeah 40% reuse and what why did you go with an adapter instead of using the wallet selector I actually dude I'll be honest I don't really understand like I don't I don't I don't understand this stuff like very well in my head like I I'm like very in tune with what's happening with the API but like like I actually don't get it and then there's this additional

84. 42:00

    thing where we're passing it through if frames and I feel like the boss on chain people like know this really well I never really understood the if frame so I I don't have a good answer to be honest yes um like Reuben Marcus meanwhile writes in the chat one is she of the new wallet selector is that some of the wallets integrated uh were still using the old newer API GS and the

85. 42:34

    wallet selector itself isn't using the latest a near API J I don't think that this is still the case uh that it's version five it's using version four so if you want uh if you want to use near aps5 on your uh application you got to have trouble so yeah and trening this uh trening confirm that this should have be one of the high priority fixes at the moment is isn't hasn't been fixed I I was under impression that it

86. 43:07

    was part of the announcements Rec yeah as far as I remember actually then by made a v to fix that to to make wallet selector compatible with four or five so that shouldn't be a problem but I don't know if it in specific packages they still hold like all dependencies that might be a problem but yeah so if anyone reports if anyone finds one please report it um I would like to get like more

87. 43:38

    takes on near API GS anyone wants who wants to uh I can say something on njs njs is a is a quite interesting Library uh I I will double down on what Mike said which is like it was not written like this on purpose and also you know like what what blad said it's not that it's like convers on and and hard to use on purpose it's just that it it does something that is hard and it went evolving right and it went

88. 44:10

    evolving with our ecosystem so there's something very interesting about near apjs that right now it could actually be simplified a lot because it used to be the only way to interact with near and therefore the only way for example to interact with a wallet and now that's not the case anymore right because a lot of a lot of these work is being done by the wallet selector right uh so there's a lot of things that we could just remove and simplify from near API JS and since

89. 44:36

    we're kind of like 10 minutes away from finishing the the talk um for anyone that is hearing I I would like to say that the hard part generally is not to find what we need to fix that's very easy you just you know like you just do yarn ad near apjs you are not whated selector you see all the problems and already you know what you have to be working on the hard part is to find someone that we all trust in the

90. 45:00

    community that can take ownership and ownership is more than just right like making a PR from time to time ownership is for example like you know Vlad can talk about it for the near C RS I can talk about it for when I took the near CLI JavaScript it's about you take the library now you're kind of alone and you have to make decisions and after those decisions you have to deal with the consequences right which is like the community will like it the community

91. 45:30

    will not maybe you forgot like at the dependency some somewhere something is broken then you have to fix it and that's why I want you know like people to not leave this com this meeting thinking like we don't know what to fix we need someone to sit down that we can all trust right and say okay you know what I'm going to take from now on the wallet selector and the near API js on one of them and then you know there's a new release ah turns out that near apjs was updated okay we have to fix it ah it turns out that a lot of people is

92. 46:04

    having problem when using the wallet selector in this place okay I'm going to fix it I'm going to review PRS we all trust that you review PRS in the right way we all trust that you're not going to you know like screw this packages because here's the you know the hardest part also like it or not near apjs is one of the most downloaded libraries that we have out there and and that's also true for wallet selector right

93. 46:29

    so it's not about maybe taking decisions it's about like right now they are in a way near is in is in a specific State just because you know it has been evolving naturally and now we need someone to take on it and make decisions and deal with those decisions and and deal with the community that that's the hard part anyway I went a little bit like you ask a question I answer another

94. 46:53

    but yeah that's my take on your ajs BL I see have your hand yeah and my very good point I would only add that it's not even about near pigs or wallet selector it's more about the developer experience and user experience so what what what makes it really hard for wallet selector is because it's not just a library it's actually a product so it's a product that end users need to use and be very efficient at onboarding so it's it's

95. 47:24

    really the frontier of near experience and um it requires a lot of um thinking and uh it's we can solve different bits and pieces here and there but the the final product will will not be there unless someone sits down and think um from the point of view of onboarding the normies the regular users uh I like liked the direction where uh near or past o went initially uh on the conceptual level

96. 48:08

    with simplifying the process as much as possible I like how near intense is currently kind of taking the another stab at it and trying to solve the onboarding experience uh themselves uh I think near AI Hub and year intents are actually much further down the line on the on that experience of making the maybe like signing uh sign in with Pass key or sign in with any wallet out there uh so they they are pushing the boundaries forward uh and it's not only about like near wallet

97. 48:48

    it's about near experience that's something I want to it but to yeah near apig GS was the the core component for developers so we can also start from that angle but we we need both um this is sort of reminding me I had a conversation with Eugene recently where I felt like I didn't really understand exactly what he was saying and I was like how about this how about since AI is so good now I will just build everything like that I understand and then if I did something wrong or you want to Pivot then I'll just put it in

98. 49:27

    Ai and we'll just do it really quickly you know so it's it's really like a different world we live in where like the planning isn't as important if AI can just like help us out and you can get it done in like 20 minutes you know so I I try to keep that in mind just like plow through get the thing working and then like pivot and who cares like that's just sort of the vibe that I roll with right now I'm a bit concerned about that direction uh when we are talking about

99. 49:51

    the um libraries and death tools because the the more pivot you have the more more users you more developers and projects you kind of uh disturb and create turbulence for sometime if if it's for no good reason then uh it may detract more than this time the reason is like fixing the entire experience with JavaScript yeah any more takes on near APS that uh you feel brave enough to share anyone any what I I want to touch on like Ruben

100. 50:35

     ruen just ru Marcus made a question which is like is there any road map in which the community could help one concern I see some dep talking is that their PRS open for more than a year with no feedb like and this is exactly what I mean about like this is exactly why why why we put the request for someone to come up and say hey I can help with this because uh is is there's simply no one to to to look at those PRS like there's something that that that is hard to convey sometimes in

101. 51:06

     the community and is like what is I'm going to speak as myself right like me gisho de head of De in particular right like there's there there's a misconception on like what are some of our roles and for example like I have a job right like which means that someone is paying me but no one is paying me for maning for example like wallet selector the same happens with trending like that's the part where we become you know

102. 51:31

     not employees of whatever company or not you know like contributors but like we become part of the ecosystem and people that care about me are in put their time like the update of the wallet selector I did not do it under a salary I did it on the weekend because I really want to see that working right like and excuse me my French uh and what is happening right now is like we just don't have like right now we are saying hey someone

103. 51:56

     we want to pay someone to please come look at the PRS make decisions and and write like and and take this to the next level uh because yeah the PRS are open for a year not because we don't care it's just that we are all doing our jobs you know it's the same way as everyone else in the community right like no one else is maintaining it for the same reason that we are not which is like that's just not my job

104. 52:20

     and I just don't have time and I do it on the weekends right and not every weekend I want to sit down and work on the award selector so to answer your question Ren Marcus the best you can do is you can go and say like if you are you know the senior that is like an expert in JavaScript and could help us then apply apply to to to help on that and if not but you know if you know someone just circulate it to them say hey did you know that these people is actually searching for someone to help uh so yeah that will be right now the

105. 52:54

     best way like find candidates and if you think that you are a good candidate please apply uh and you know like for anyone also listen to this you're not going to be in the boil like you're going to start with most probably some of us helping you the the whole idea is that after some time we would really like you know to let your hand go and you just keep doing the things in the right

106. 53:19

     way and just to support this idea I want to bring a different example how that works already in the rust St so we now have the new revamp near API RS Library which is also experimental and people are like bringing their own feedback but mostly it works and there is a clear ownership and the the author Arthur who's not on the call I believe today with us uh but he is very active now maintaining a lot of rust tools uh in

107. 53:52

     your ecosystem gets totally fair uh compensation through through devhub um so it's up to the cool so and there is a question in the chat that I'm reading right now uh will be there a position description anywhere um so the there is no one to to to interview the person to be honest but at the same time we have the so the the model is a bit different that we are operating it uh so there are RFP request for proposals and the proposals that people can uh

108. 54:32

     put there I know that's different from employment uh so it's not for everyone but uh we are exploring this direction yes yeah and while we are on this topic since uh Vlad you're representing not only near def but the infrastructure committee uh it's um like is is there any plan if for example we don't find or at least our community does not have the capability of taking this ownership at the moment for be for wallet selector or near API JS um would

109. 55:12

     it be viable of trying to outre to to reach outside of our community like a different teams I know that we have had uh um like different teams taking or supporting to an extent uh some of those libraries and how can we as a community um try to enhance that that we are searching for maintainers I can share the experience searching for teams that were not really interested in building stuff but they were interested in just getting the money um that didn't play

110. 55:50

     well and and it it requires a lot of um then management from our side or from whoever and but we are actually looking for a p for a person to internalize and learn the challenges pass it through themselves and come up with upated design that would um you know reflect their vision and drive it and we are offering the finding for that so I just wanted to agree with lot about the idea that you cannot build the good

111. 56:30

     product if you don't use and you don't don't use it if you don't feel the pain of the regular user if you don't like meet this pain every day you cannot efficiently Sol it so this is what I believe so definitely you need to have the idea around know the pain of the people of the teams and only after that you can you can Sol it this is what I think about this yeah I want something I want to add something on top of that because what what you said is not minor I I I I I I I

112. 57:09

     believe that what you say is very much true and something else like something that we found when we were trying to to to get people from you know like you just go and hire a company to fix the problem something that you will find you know pretty much soon is that if they don't actually you know care about the community or if they're actually not suffering also the pain points they just don't understand what what is important or not and also they never take the

113. 57:33

     initiative of saying like ah and you know what I saw also that bet is trying to do this and therefore we could head like this or you know like now the near intense is coming maybe we could just be proactive and at this other thing and and that's what I was saying also before about like finding someone is not just you know like having ownership is not only you know fixing PRS because that's you know that's doable it's also about like making decisions like you know

114. 57:58

     believe it or not at some point is not telling me what am I supposed to be doing with documentation I just go look what the community needs and what other people need and I and I just go and do it and we will need exactly the same right like someone that not only suffers but also cares so much and so deeply about the community that they take proactive actions to make the user experience better and that's what makes

115. 58:21

     so hard to to you know find someone amazing and since we just about crossed the hour uh I would like um to ask if anyone of you have any closing thoughts otherwise we we can close this session off and have a little bit of thought of how we can not only proceed but collaborate further so if any of you have final thoughts I don't have a final thought but we have a final uh action item call uh for everyone is to go check out the

116. 59:02

     rfps on the infrastructure committee I hope lill will include the link in the uh video recording below and um there are several projects that actually are promoted there uh they are more or less interconnected uh at this point we you can name them as on boarding experience and you can be the owner so consider doing that um looking for teams uh so uh thanks a lot thanks everyone for being here for those of you watching once again if you want to present your work suggest discussion topics or want to go get involved join

117. 59:46

     our telegram group uh this was the tool call uh and we have those every second Tuesday of the month so if you want to present join us uh on the next one otherwise I would like to thank everyone of you for uh joining this call and being here and I hope to see you next on the see you on the next one thank you thank you guys see you