1
00:00:00,000 --> 00:00:14,000
Absolutely. First of all, this is all for Derek and Semisow really to actually develop the nest itself and to fund Semisow.

2
00:00:14,000 --> 00:00:21,000
And we have a tip jar on the top here. Here you can see the donate button with the frog emoji.

3
00:00:21,000 --> 00:00:32,000
So feel free absolutely to donate to the fundraiser really. Let's get the nest as good as they can be.

4
00:00:32,000 --> 00:00:38,000
Yeah, in the meanwhile, I think Sleep is going to make a post so we have everybody in the audience

5
00:00:38,000 --> 00:00:52,000
chime in and everybody know that the first time lost time slot is going on. But I'm really, really glad to have you Mazen and Katie on. This is absolutely incredible.

6
00:00:52,000 --> 00:00:59,000
Hey, yeah, thanks for having us. Yeah, we're excited to be here.

7
00:00:59,000 --> 00:01:23,000
Excited to have you guys Mazen. If you like, can you please like, explain to us a little bit about relays for any of the people down in the audience or anyone listening to the recording, and they might not be knowledgeable on what a relay is.

8
00:01:23,000 --> 00:01:27,000
Can you can you please explain a little bit Mazen. Thank you so much.

9
00:01:27,000 --> 00:01:29,000
Sure.

10
00:01:29,000 --> 00:01:41,000
I mean, I'm definitely not a relay implementer, super genius or anything but relays very basically are what store your notes and other stuff.

11
00:01:41,000 --> 00:01:58,000
And basically, generally they use some form of database structure with indexing. And right now the the fastest implementation and the one that we basically are using across all of our projects is stir fry.

12
00:01:58,000 --> 00:02:15,000
And thankfully, Doug is the one who made it, because I think Nostra would really be struggling without him today. So I think I mean that at a very basic level is all relays really do they accept requests from clients, and they respond with events.

13
00:02:15,000 --> 00:02:28,000
And when you send your events to them they basically store them and later accept requests for that content. Did I miss anything.

14
00:02:28,000 --> 00:02:31,000
That's awesome. That's awesome.

15
00:02:31,000 --> 00:02:47,000
Yeah, so you guys run the the wine relay and I am so thankful that I found it, because it's been so good to have pure signal on the global feed. It's been amazing. I'm a really really happy customer.

16
00:02:47,000 --> 00:03:06,000
And I saw that that that I got this reply from a bot got sent straight to my DMs, letting me know that my subscription was almost over. And it was very simple because they sent me an invoice and I was able to just pay right there and then and knew my amp up and everything was amazing.

17
00:03:06,000 --> 00:03:10,000
I love the way that you guys did that.

18
00:03:10,000 --> 00:03:19,000
So yeah, I just wanted to ask you, what are your plans for for the foreseeable future with wine, and the subscription model.

19
00:03:19,000 --> 00:03:23,000
If you could chime in on that. Thank you.

20
00:03:23,000 --> 00:03:28,000
Sure, I want to clarify first because we have, we get this question, all the time.

21
00:03:28,000 --> 00:03:37,000
And we haven't moved Nostra.wine to a subscription. The only thing we've moved to a subscription is filter.nostra.wine. That's it.

22
00:03:37,000 --> 00:03:49,000
And the reason is because filter.nostra.wine is the aggregator and the broadcaster, and it is a lot heavier of an operation for us to run that required a lot of custom code that we implemented.

23
00:03:49,000 --> 00:04:01,000
Nostra.wine, a high performance, regionally mirrored, great paid relay, but for now we don't have any plans to change anything about the fee structure of that. It's still one time.

24
00:04:01,000 --> 00:04:06,000
The only thing that we've added a subscription to is filter.nostra.wine.

25
00:04:06,000 --> 00:04:26,000
But we actually have, I have a little bit, a few thoughts that for kind of this talk, and so we can kind of direct a little bit of this early stuff and then when we run out of things to talk about we'll let you guys just bombard us with questions about kind of whatever you want if that's, if that works for everybody.

26
00:04:26,000 --> 00:04:28,000
Yeah, definitely, definitely, definitely.

27
00:04:28,000 --> 00:04:47,000
Okay, so I kind of wanted to start with like how I got to Nostra. I've course heard about it from Lightning and stuff but I wanted to start a relay because basically we've always run both Katie and I kind of like big applications at scale.

28
00:04:47,000 --> 00:04:58,000
And we like client facing applications that's kind of like what excites us. So a relay was an opportunity to kind of pursue like a big data project that scale.

29
00:04:58,000 --> 00:05:07,000
And we knew that eventually we could probably add value there as well. But basically I set up a public relay that's the first thing I did.

30
00:05:07,000 --> 00:05:28,000
Using no stream so shout out to Khmary for providing no stream to the community. And, you know, within a week, there was just an unbelievable amount of spam. And at the time nobody had really come up with many good ways to interact with spam control on a relay level.

31
00:05:28,000 --> 00:05:37,000
You'd have to basically write everything custom there was no way to basically get a stream from the relay and decide what to do with that event.

32
00:05:37,000 --> 00:05:50,000
So, spam was basically a non starter for public relays in those early days and I kind of knew there's not a sustainable model with running a public relay, you're basically offering a public service is going to be a donation.

33
00:05:50,000 --> 00:06:05,000
And I think some people will do it and I'm super thankful for them. But the incentives for us to run something sustainable made like a lot more sense so when I found out we can move to a paid relay that address the spam problem and potentially made the operation something

34
00:06:05,000 --> 00:06:19,000
that I could run for the foreseeable future. That was, that was the direction I went in right away so I kind of set up Nostra.Wine. And Katie came along somewhere in the middle of that I had kind of been chilling her Nostra for a little while and she's like

35
00:06:19,000 --> 00:06:25,000
she's not too big on social media so she was kind of like I don't really know what I need that for.

36
00:06:25,000 --> 00:06:36,000
But once she kind of saw the experience that I was having, notably on on Domest because that was, I kind of got in with that February rush with everybody else when it hit the main app store.

37
00:06:36,000 --> 00:06:54,000
She kind of, she understood what I saw and how easy it was to work with the protocol. And once she kind of got involved and got on Nostra.Wine, and the domain was her idea by the way so I think it's a really great domain and we're drinking wine right now.

38
00:06:54,000 --> 00:07:10,000
But once we once we got that domain up and running.

39
00:07:10,000 --> 00:07:27,000
I tried to remove them from my list I only use paid realize, and I got like 1012 15 I basically paid for every paid relay that existed in the first like two weeks of that client, the paid Nostra implementation getting released.

40
00:07:27,000 --> 00:07:40,000
I tried to adding all those and removing all the public relays but the problem is I still had a ton of missing content, because there just are a lot of Nostra users who aren't going to pay for realize.

41
00:07:40,000 --> 00:07:47,000
And worse than that, I thought was like, now my contents not even getting out to these users because a lot of them weren't even reading from the paid realize either.

42
00:07:47,000 --> 00:08:05,000
So that kind of led us to creating filter, which really architecture of filter is is Katie's brainchild, if you'd like, I'll let you kind of talk about, you know, the origins of filter how we put it together and kind of simple terms, and then, you know, we can

43
00:08:05,000 --> 00:08:19,000
kind of talk about the other advantages that emerged that we didn't really think of initially, and then kind of our future plans. Yeah, so I mean filter like you said it really came from things that we were looking for in a product ourselves.

44
00:08:19,000 --> 00:08:40,000
The paid only relay solution for us was missing a lot of content it was missing the broadcast feature. And so this idea of just being able to augment the global experience, but maintain kind of that rich public experience on threads and on mentions

45
00:08:40,000 --> 00:08:53,000
and everything else. So filter really was kind of an idea of just trying to augment that. And there's a lot. Now that we've built it in terms of what we want to do with it in the future.

46
00:08:53,000 --> 00:09:03,000
It's funny we have a story where we really did the augmenting of the global first and I remember while working on it and while testing it.

47
00:09:03,000 --> 00:09:12,000
Man we really need broadcast man we really need broadcast but that wasn't a feature that we went out with first and I think shortly after releasing it.

48
00:09:12,000 --> 00:09:30,000
Somebody mentioned Hey, can we broadcast on here. And we added that in that night because I just couldn't stand not having it so I mean it really it's kind of a product that really came out of just us and our things that we wanted with it so in terms of where I kind of see it

49
00:09:30,000 --> 00:09:51,000
going. And where I'd like it to be is really a tool that enables for high level of user configurability with off that actually allows for a lot more user configurability than we had with the parameter, you know style before which was really clunky.

50
00:09:51,000 --> 00:10:10,000
So now we are we're able to off users, and really we can maintain user level configurations and so where filter should go or where I think we're going to take it is really starting to allow for lower level configs I know as an extended filter to allow for the true global now

51
00:10:10,000 --> 00:10:14,000
that we started to have some basic spam filtering.

52
00:10:14,000 --> 00:10:25,000
I think there's, there's even farther we can go the way that I talked about it a lot is really when you think of like Apple in their focus modes.

53
00:10:25,000 --> 00:10:40,000
The content that you want at any given time it's not even just user level it's user mood level or user activity level, if I'm working, I might want something different than in the evenings or if I've had a stressful day maybe my evening looks a little bit different

54
00:10:40,000 --> 00:10:55,000
and I want, you know, just to look at art and music or something a little bit more low key so you know as a user, what I want in a feed really does look different from day to day or from moment to moment and so long term.

55
00:10:55,000 --> 00:11:11,000
I'd like filter to be able to support that level of configurability without having to support necessarily multiple independent relays, so this gets into another was a question around machine learning and what we're going to do with machine learning.

56
00:11:11,000 --> 00:11:27,000
Right now we've been focused a lot on the architecture of filter the architecture of some of the other things that we've been building in the data engineering pipelines to layer in machine learning that's really usable.

57
00:11:27,000 --> 00:11:42,000
We're going to need some really robust data pipeline so that's really kind of we've been working on the foundations of a lot of this but where we want to take it is enabling topic modeling and classifications of posts really at interest level, you know,

58
00:11:42,000 --> 00:11:57,000
we're not talking about music this is about art, or, or even, you know, a little bit more generic than that, but really labeling the data enriching the data with labels, making that completely transparent because I don't think that user configurability should come at

59
00:11:57,000 --> 00:12:12,000
the expense of, you know, censorship like we want to make sure that those labels and how we get to them is clear that there's feedback on those labels if they're ever wrong, but really trying to create tools for user users to create their own experience.

60
00:12:12,000 --> 00:12:25,000
So that's really the way filter setup, is it set up to be able to kind of grow and support more, you know, dynamic user configurability in the future.

61
00:12:25,000 --> 00:12:40,000
Yeah, just like to give a couple examples, and she mentioned, you know, being in a maybe not as great of a mood and only wanting to see specific content but you can get really granular with the filtering depending on how good the classification is so like one of the things

62
00:12:40,000 --> 00:12:42,000
I love the true global feed.

63
00:12:42,000 --> 00:12:56,000
I actually do really enjoy that this man filtered one obviously when it's just all repetitive spam that's not very good and the only spam filtering we do just be clear is, is the intense repetitive messaging that I think there's pretty good consensus that nobody wants that

64
00:12:56,000 --> 00:12:58,000
right now.

65
00:12:58,000 --> 00:13:13,000
I love the the pure global feed but my problem is, there's too many bots, and we're not doing any filtering with those because people like to interact with them and and bots are a good thing for no stir, but I don't want my entire global feed to be like 80% bots

66
00:13:13,000 --> 00:13:16,000
which is basically what it ends up being.

67
00:13:16,000 --> 00:13:31,000
Bots can be labeled and classified as well. And so I could have a user configurable filter that is as simple as saying I want the pure global, but remove the bots like that's something that I desperately want today that as far as I know doesn't exist.

68
00:13:31,000 --> 00:13:49,000
So this can offer some filtering like this but what we found is actually the whole idea of smart clients and dumb relays isn't a one size fits all thing, we think smart realize have a huge place, because if you only have dumb realize and you have

69
00:13:49,000 --> 00:14:01,000
a client level they still need to do a ton of like post processing when they receive the data deduplication they have to parse out things if you want to have filters on the client level that's going to slow down your clients.

70
00:14:01,000 --> 00:14:15,000
So, having distributed, but high performance realize that do some of that work for you is a part. You shouldn't be the only thing but it is a part of like a very healthy decentralized ecosystem, in our opinion, at least.

71
00:14:15,000 --> 00:14:29,000
So the next step we want to go is continue expanding user user configurable, we're basically calling them focus modes that's kind of our concept because we want each user to be able to have multiple.

72
00:14:29,000 --> 00:14:44,000
And then, you know, there's a bunch of other advantages that we didn't really have in mind when we started filter that kind of emerged and like one of one of the biggest ones people learn pretty quickly using Nostra on mobile is the bandwidth use can be insane.

73
00:14:44,000 --> 00:15:01,000
And as you add more and more relays, you basically just increase that over and over and over again because you're just adding way more duplicates. And I used to think that like all this bandwidth usage is just image rendering and don't get me wrong a lot of it is, but actually

74
00:15:01,000 --> 00:15:14,000
the amount of bandwidth that the relays use is pretty mind blowing. So, if you think about the amount that your client is pulling from I've seen people you know with 7080 100 relays.

75
00:15:14,000 --> 00:15:33,000
It's an unbearable amount of data that that most mobile plans and speeds really can't shouldn't be dealing with. So, bandwidth less bandwidth use is a huge benefit that we got out of it and higher performance and faster response time with with removing

76
00:15:33,000 --> 00:15:49,000
duplicates basically that's a huge part, and then something that we're pretty excited about but we think is pretty underused right now in the protocols is we did implement search on filter as well and shout out to our turn his work on Nostra band and nip 50.

77
00:15:49,000 --> 00:16:03,000
Because he basically did, as far as I know he did almost all the work on that and and made basically the only implementation for a very long time. And he basically carries the entire amethyst search function on his back right now.

78
00:16:03,000 --> 00:16:14,000
So we added that to filter so we want to support that for our customers and we could add it to to Nostra dot wine and we have kind of thought about that and talked about it.

79
00:16:14,000 --> 00:16:23,000
But the reality is not sure that one is only has the content from its paid users and filter has the aggregation. So the search is much more powerful on filter.

80
00:16:23,000 --> 00:16:34,000
And we exposed it via a Kibana instance which is basically just like a website where you can explore elastic search, and it's searched out Nostra dot one.

81
00:16:34,000 --> 00:16:41,000
And I don't really think very many people use it but I use it every day I think it's pretty cool, you can get pretty granular on your search of events.

82
00:16:41,000 --> 00:17:06,000
And we dropped a lot of the kind of sensitive stuff because we know you can get to it but we don't really want to be the people that offer DM metadata to everybody and things like that so we got we got rid of like a lot of events that we don't think are very valid for search, but all the kind ones and zeros and threes and zaps and everything in there from at least like I think it's been about a month that we've been aggregating from there so

83
00:17:06,000 --> 00:17:09,000
anything else you have about filter.

84
00:17:09,000 --> 00:17:23,000
Okay, so I guess we'll take a pause there and say like does any do you guys or does anybody else have questions about filter or future plans for filter before we move on to other topics.

85
00:17:23,000 --> 00:17:29,000
I think, honestly, we just mentioned about search got me super excited.

86
00:17:29,000 --> 00:17:46,000
Right there and then I'm trying to comprehend exactly like, like, all the capabilities of what I can actually do with it. And something that came to mind was, I sometimes cannot find a specific no right so I would have to go over to national band.

87
00:17:46,000 --> 00:17:59,000
And there I could see everything I was trying to remember right now. If, if it would sort of be something like that when I could, could it be by date, can I search by date alone.

88
00:17:59,000 --> 00:18:03,000
And, or is that something I might think about something else.

89
00:18:03,000 --> 00:18:12,000
Yeah, you can search. You can definitely search, you can make as, as, I mean pretty much as complicated as searches, you want.

90
00:18:12,000 --> 00:18:31,000
So Kibana doesn't support quite as complex as everything as the back end does but absolutely you can chain together as many requirements as you want so you could say I want this ID created at or since this timestamp with this additional content filter,

91
00:18:31,000 --> 00:18:35,000
you can add as many string together as many filters as you want.

92
00:18:35,000 --> 00:18:52,000
The relay implementation doesn't allow you to do all of that because that's not what's in the spec the spec is basically just keyword search, but you can apply that keyword search to additional things like kinds or IDs and other filters so it's not quite as granular

93
00:18:52,000 --> 00:19:00,000
as we can support on the web and through our API we also offer the search through an API docs dot noser dot line if you're curious about that.

94
00:19:00,000 --> 00:19:01,000
Oh yeah.

95
00:19:01,000 --> 00:19:16,000
But yeah the relay implementation is a little bit less advanced just due to the structure of the spec but there will probably be improvements on that spec and we'll improve the relay implementation when those come.

96
00:19:16,000 --> 00:19:21,000
That's awesome. That's awesome. As long as we have keywords. I'm super happy with that.

97
00:19:21,000 --> 00:19:31,000
That's a yeah that's that's all I wanted to really yeah I also have some questions coming in through the NOSTER chat so I'll save those for later.

98
00:19:31,000 --> 00:19:36,000
And then we can open up a live Q&A if you'd like.

99
00:19:36,000 --> 00:19:45,000
Okay, yeah if there if there are questions that are particularly relevant to stuff we just mentioned you can feel free to ask them too because it might be easier to stay on the same topic.

100
00:19:45,000 --> 00:19:54,000
This one isn't very it's only it's only about he's saying what is an ideal amount of quality release for the average user.

101
00:19:54,000 --> 00:19:56,000
Okay, yeah.

102
00:19:56,000 --> 00:20:03,000
Yeah, we could have saved that but I'll answer just because it's quick I don't I don't have the answer actually so that makes it really easy.

103
00:20:03,000 --> 00:20:14,000
I use basically just our relays because it's easy for me to trust myself. It's not a big deal. That's not a setup that I generally recommend to other people because I think having a little diversity in your setup makes sense.

104
00:20:14,000 --> 00:20:17,000
For a lot of reasons.

105
00:20:17,000 --> 00:20:21,000
But I think the number is a lot less than some people think.

106
00:20:21,000 --> 00:20:34,000
And but it's it's it's ultimately up to you if you're using NOSTER mostly on your desktop computer and you have unlimited bandwidth, and you want to use 100 relays and you your client experience is okay.

107
00:20:34,000 --> 00:20:48,000
I say go for it. For me the number is below 10. That's where that makes me feel like the client experience is fast. And I'm not basically my experiences is the only little extra tidbit I'll give you.

108
00:20:48,000 --> 00:21:01,000
I don't have a perfect network view I think people like NOSTER band and and Blake and some of the stuff he's doing with NOSTER graph they kind of have a better total network view because our goal is never to aggregate every relay on the network that's not we never wanted to do that.

109
00:21:01,000 --> 00:21:12,000
So, but I have a good view because we basically look at like the top 10 to 15 relays, and I found once you get out of the top three or four, it's almost all duplicates.

110
00:21:12,000 --> 00:21:16,000
So if you think if you're loading 100 relays.

111
00:21:16,000 --> 00:21:33,000
I mean, yeah, I'm not sure that you're advancing any purpose, besides getting away more duplicates. So that's a short or way too long answer. The only thing I would add is that I do think we're going to see the emergence of a lot more specialized relays and community

112
00:21:33,000 --> 00:21:46,000
relays, and in those instances, there's just less of that issue of duplication. And so, the answer today may not be the same as the answer in a month or two.

113
00:21:46,000 --> 00:22:05,000
And depending on what types of relays, you're, you're using, it may not be the same so I think it's really about how much redundancy. You're comfortable with, and then access to any specialized content that you know may not be available on some of the public

114
00:22:05,000 --> 00:22:10,000
relays or bigger relays.

115
00:22:10,000 --> 00:22:21,000
Yeah, I think that's, I think that's good, even better than what I said. So, yeah, that could change depending on like the specificity of relays which I think is coming.

116
00:22:21,000 --> 00:22:36,000
One last thing I want to touch on because I think probably most people in here are aware of it. But one thing that having filter in the aggregator that powers it allowed us to do is make the welcome dot noser dot wine relay, which I'll be honest, I didn't know if

117
00:22:36,000 --> 00:22:46,000
it's gonna work it's, I honestly made it one day on vacation, Katie's been giving me a lot of shit for it actually because she's like why would you make this on vacation.

118
00:22:46,000 --> 00:23:00,000
But I did it, and I got super excited about it while I was working on it because there were so many, there was some boost that like two weeks of new users not we've had several of them on the sir but that was one of right when I released that relay

119
00:23:00,000 --> 00:23:08,000
there was this like binge of new users coming in. I think it was because the Dogecoin thing on Twitter actually.

120
00:23:08,000 --> 00:23:24,000
And having a relay where we could see those people coming in and talk to them right away, I think is making a difference. I know our retention is still probably not great, but one of one of the things that's challenging about like a decentralized

121
00:23:24,000 --> 00:23:40,000
protocol where all the clients are going to be different and there's not going to be perfect like uniformity, and it's not going to always look the same as what people are used to. It's, it's really important to have people help you, or to have really good onboarding process that that helps you out so right now

122
00:23:40,000 --> 00:23:47,000
until we've kind of perfected the onboarding which is like very complicated I don't think anybody has a good solution for that yeah I certainly don't.

123
00:23:47,000 --> 00:24:02,000
Having a welcome relay that basically just isolates new users, new user posts is is a very useful thing that we can can help people welcome people to the community so the little benefit of filter that I definitely didn't have in mind when I built it but I was like

124
00:24:02,000 --> 00:24:08,000
hey we have all this data now we can actually track when it's the first time we've seen a new user.

125
00:24:08,000 --> 00:24:23,000
I don't know if I cut off there at the end but anyway, though, the other thing that we want to talk about and I don't know, I guess it's a good time to get into that is kind of what we've been that our next thing, and we're obviously not going to stop developing

126
00:24:23,000 --> 00:24:31,000
filter that's still kind of our, our gem, but there's something that people have been asking for a lot.

127
00:24:31,000 --> 00:24:44,000
For as long as I've been around. And I think there's a lot of ways to do it but offering a relay that helps with it is something that we're interested in. Do you want to talk about it.

128
00:24:44,000 --> 00:24:53,000
Okay, so we, we have basically been developing what I'm calling a Patreon style relay.

129
00:24:53,000 --> 00:25:18,000
The idea is that we'll be able to onboard creators who can create tiers of basically subscription amounts that they want to pay wall any kind of content that can be journalists that can be artists that can be only fans, it can be whatever the users want it to be.

130
00:25:18,000 --> 00:25:39,000
So, the idea will be will onboard creators will provide content and file hosting for our onboarded creators, and then we'll provide a relay where users can come on subscribe to any number of creators they want one or many, and their feed that they get in global

131
00:25:39,000 --> 00:25:56,000
and in following and whatever else will be only the content that they've subscribed to for the creators that they've subscribed to. So that's the general concept that basically allows individual subscription relays for users except with us doing all the hosting.

132
00:25:56,000 --> 00:26:11,000
And part of this goes back to onboarding and trying to bring in new, new people into this ecosystem. And so I think one of the things we've seen is there's a lot of artists and different people trying to create content in the community and we really wanted to give them

133
00:26:11,000 --> 00:26:32,000
a place where they could get value for value for that and so this enables them to really have like a clear way to take advantage of the protocol and monetize their, their content, while also trying to, you know, offer them some protection for that content to be easily

134
00:26:32,000 --> 00:26:39,000
shared and kind of just some kind of gatekeeping around that just to allow them to monetize that.

135
00:26:39,000 --> 00:26:51,000
So one of the struggles right now and even if we get a good which will absolutely implement but even if we get a good like zap paywall protocol level kind of spec if someone comes up with a good way to do that.

136
00:26:51,000 --> 00:26:57,000
You still kind of have the problem where if you host your content on a public image host.

137
00:26:57,000 --> 00:27:13,000
And if you don't have that link it's really out in the wild, right, and so that that's kind of the struggle that everybody deals with so our ideas to provide the hosting ourselves, then we can do some really granular control on how long individual users have access

138
00:27:13,000 --> 00:27:29,000
to the content so that ultimately the link we give you, if you go and put it out on the internet people won't be able to use it so it's it lets us do a lot of the kind of creator level protection, it centralizes some of obviously the file hosting and things

139
00:27:29,000 --> 00:27:44,000
that creators are welcome to bring their own file hosting and do whatever they'd like with that, but we can't control the content access if they do that so that's that's kind of the advantage that we're going to try to to bring.

140
00:27:44,000 --> 00:27:47,000
What did I mess.

141
00:27:47,000 --> 00:27:50,000
I think that's, I think that covers it.

142
00:27:50,000 --> 00:27:59,000
I think that's what happens about kind of that concept because that's basically been what we're working on we're pretty close. I don't want to give like a exact timeframe.

143
00:27:59,000 --> 00:28:11,000
But we, we worked on it all day today and we're a few weeks into working on it now, and most of it is is put together it's just kind of ironing out the last parts there's a lot of moving parts actually to make it work well.

144
00:28:11,000 --> 00:28:20,000
And I was, I was laughing earlier because we have like something like 2000 lines just the back end code for this project and we had never run it.

145
00:28:20,000 --> 00:28:34,000
So we ran it for the first time today and it mostly runs, so we're pretty thrilled about that and then we just need to sort out a couple other things and we'll be, we'll be able to launch that and so one of the things we want to tell you all is that if any of you

146
00:28:34,000 --> 00:28:46,000
are content creators of any kind, we don't really care what your content is, there will be kind of different rules, depending on what your content is but if you're a content creator of any kind, and you have any type of following.

147
00:28:46,000 --> 00:29:01,000
Send me a DM, and we'll be able to hopefully start onboarding people in the next week or two, and we want to kind of do a small beta initially and and troubleshoot some of the issues because we know there'll be some of those, but once we get this ironed out you.

148
00:29:01,000 --> 00:29:13,000
We want to make sure that you have the same type of performance and right reliability that you get from the rest of our relays here too.

149
00:29:13,000 --> 00:29:24,000
And yeah, I think that's about all we have to talk about we can answer questions now I can look at some of the questions from the threads but those are just kind of the main stuff we have to address.

150
00:29:24,000 --> 00:29:38,000
Yeah, I can't wait. I'm curious to see how the tiers are going to work for each individual content creator, and what they're going to think of what they're going to think about offering.

151
00:29:38,000 --> 00:29:41,000
It's going to be crazy. I can only imagine.

152
00:29:41,000 --> 00:29:46,000
I've already seen I've already seen a crazy demand for it.

153
00:29:46,000 --> 00:29:51,000
Yeah, I know exactly. Yeah, I can see a lot of people benefiting from this.

154
00:29:51,000 --> 00:29:54,000
That's amazing.

155
00:29:54,000 --> 00:29:58,000
Do you have anything to say about that specifically.

156
00:29:58,000 --> 00:30:12,000
Personally, I have nothing to really add I'm just really really excited for the beta to be honest and to see what how I can maybe onboard a lot of like content creators outside of master and everything to.

157
00:30:12,000 --> 00:30:22,000
I think that would be really good as a lot of the people are trying to find a way to express themselves and maybe that could be a great way to do that.

158
00:30:22,000 --> 00:30:26,000
Who do you have this end up, do you want to come up on stage.

159
00:30:26,000 --> 00:30:39,000
Give it a give it a quick second, and let's let's go to like the 45 minute mark, and then we can open it up but I could bring in questions from from NASA chat if there's any.

160
00:30:39,000 --> 00:30:51,000
Yeah, sure. I'll do one other thing as well while you're waiting like one of the advantages kind of that we can offer over those centralized performance platforms I think there's a couple advantages right like we're going to do some centralization on the

161
00:30:51,000 --> 00:31:01,000
relay and the content hosting side because it is what it is there's no way for us to control the accessibility of it unless we do that and I think that is what creators are looking for.

162
00:31:01,000 --> 00:31:10,000
They are still free to post as much of their content free for zaps or whatever as they want to but if they want to be behind a subscription it kind of has to be like that.

163
00:31:10,000 --> 00:31:25,000
But what you get on Noster that you don't get in most other places or Noster I've been saying no strength I'm going to get in trouble. I think, I think Noster is preferred these days but we're big Noster people around here so

164
00:31:25,000 --> 00:31:34,000
I'm going to have to poop at me oh my god. Okay, so, but where was I don't even know what I'm talking about anymore.

165
00:31:34,000 --> 00:31:48,000
So, we get what we get is interoperability with clients that's like the biggest value prop that we get from Noster. So, no matter how you want to consume your content that you subscribe to.

166
00:31:48,000 --> 00:31:55,000
You'll have to go on various clients, you can create your own client experience that gets you there.

167
00:31:55,000 --> 00:32:07,000
So, I can imagine like some creators will be more image and video focused there may be clients that don't even exist yet today or clients that do exist today that are way better for viewing that stream.

168
00:32:07,000 --> 00:32:13,000
And because it's just a relay that anybody can use you'll get that interoperability kind of right out of the box.

169
00:32:13,000 --> 00:32:22,000
The second big benefit that we can offer over the centralized platforms is since we don't have to deal with the dirty fiat processing.

170
00:32:22,000 --> 00:32:39,000
The secondary kind of zaps that the creator gets beyond the subscription are theirs and theirs alone and get paid directly to them. So there's no need for us to get involved in the secondary payments that are going directly to the content creators,

171
00:32:39,000 --> 00:32:51,000
which is a huge additional value I think for them because if you ask, I would assume a lot of these content creators make more money post subscription than just the subscription.

172
00:32:51,000 --> 00:33:14,000
So it's kind of a lot of thoughts around those and being able to manage all that in fully customizable to each creator and letting the users get value out of it through the interoperability of clients and being able to use one relay for all the creators you want to subscribe to no matter what their genre or product is.

173
00:33:14,000 --> 00:33:16,000
That's incredible.

174
00:33:16,000 --> 00:33:32,000
I have a question here from Ethan. So Ethan says, What kind of changes do we need to make on the client side to make relay management easy and efficient for relay operators. Also, how should we index old events.

175
00:33:32,000 --> 00:33:36,000
You have any thoughts on this, or you want me to start.

176
00:33:36,000 --> 00:33:54,000
Well, I think part of this is there's there's been, I know one topic is probably really niche that's come up a lot from clients and relay operators is there is a limitation I think that a lot of relays have on the number of active subscriptions that can take place within a given

177
00:33:54,000 --> 00:34:12,000
time frame. And I think this is just an opportunity for relay operators and client developers to communicate because if we as really operators limit that count too much, we start encouraging bad behavior on the client side in terms of chunking or collecting filters to a subscription

178
00:34:12,000 --> 00:34:31,000
ID that may not have the same lifecycle. And so some of this is a push pull. I'll say some of this is about relays understanding appropriate limits as well as, you know, client behavior. So I don't want to put all of the, I think I guess it's a working relationship on both sides.

179
00:34:31,000 --> 00:34:33,000
It's really my comment.

180
00:34:33,000 --> 00:34:43,000
Yeah, I would say the, the biggest thing they could do is try to be efficient with their requests, obviously, I mean, and I think everyone's trying to do that because it affects client performance too.

181
00:34:43,000 --> 00:35:00,000
But there definitely were some early clients that were like very very absurd about the amount of requests they'd send and keep open. And there's also some clients that seem to open an unbelievable amount of connections that we've never totally figured out why.

182
00:35:00,000 --> 00:35:14,000
But one of the things actually, we can give a little advice to relay operators that that we've implemented is filtered just kicks you off after an hour, you know, and you need to reconnect back.

183
00:35:14,000 --> 00:35:23,000
And for most clients it's it's not a big deal at all because they auto reconnect and every time you like kind of move to a new page there's some type of cycle where it kicks off those connections again.

184
00:35:23,000 --> 00:35:35,000
But the reality is it cleans up so many stale connections. We don't have that many filter users and there were times where I would see like 10,000 connections to filter, it just doesn't make any sense.

185
00:35:35,000 --> 00:35:51,000
And ever since we implemented the timeout restrictions, you know, that number stays or stays, you know, at a way more reasonable level basically so adding timeouts to the lifecycle of your web sockets is not a bad thing, at least we don't view it as a bad thing.

186
00:35:51,000 --> 00:36:02,000
And, and most clients aren't on the client for an hour straight without like leaving the app. If you think about like most mobile clients, you're disconnecting from your realize every time you you tab out of the app.

187
00:36:02,000 --> 00:36:12,000
So, an hour timeout really affects very few people and mostly controls, like leaky connections that haven't been closed.

188
00:36:12,000 --> 00:36:20,000
Oh, wow, I've noticed that I'm in all having to populate all over again and taking some time every time I refresh the app.

189
00:36:20,000 --> 00:36:34,000
Yeah, if you if you leave most of the mobile clients if you leave and they drop they're not working in the background basically so that's why you like people don't have notifications out of the box right now because we're not working in the background so they're connecting back to your

190
00:36:34,000 --> 00:36:40,000
relays once you open back up the app so there's always going to be like kind of a little initial delay.

191
00:36:40,000 --> 00:36:47,000
Okay, that makes sense. His second question I believe was, how do, how should we index all the vents.

192
00:36:47,000 --> 00:37:01,000
So that I guess that depends on what your purpose is so we're using elastic search for search, of course, because we're not trying to reinvent the wheel and they do it so incredibly well.

193
00:37:01,000 --> 00:37:12,000
So for search, that's an out of the box solution that's like very easy and supports JSON and there's not much to make it worse if you're trying to run a relay.

194
00:37:12,000 --> 00:37:29,000
Okay, that's a different question. I think he's asking because he's actually, he has this on this thing I think it's micro app, where he actually stores your, your events, like through, like he aggregates all the events that relates also I guess

195
00:37:29,000 --> 00:37:37,000
and he tries to like save them for you. I'm gonna try to find it the link real quick. Yeah, like a backup service. Yes, exactly. Okay, that makes sense.

196
00:37:37,000 --> 00:37:49,000
And the one thing I was, it's not, I don't believe it's going to be a one side one size answer for everything like if you're building a relay that's specifically for archiving.

197
00:37:49,000 --> 00:37:56,000
There's going to be optimizations around indexing that's specific for that I don't know that we've thought specifically about that problem, we haven't.

198
00:37:56,000 --> 00:38:07,000
There will be a time when we're going to have to archive a lot of data and I personally, as a data addict don't want to lose it. So we will be thinking about that.

199
00:38:07,000 --> 00:38:25,000
But yeah the answer for indexing and storing data really should be optimized based on the use case, access requirements for that data. Yeah, so it's, it's going to look different for some relay that's trying to operate as an archive

200
00:38:25,000 --> 00:38:32,000
relay versus a relay that's trying to operate as, you know, a source for live events.

201
00:38:32,000 --> 00:38:42,000
Yeah, that's really the only answer. Yeah, we haven't done a ton of additional archiving outside of our relays because one of the nice things about the way we've set up.

202
00:38:42,000 --> 00:39:03,000
Not sure that wine is it's already, there's, we already have regional mirror relays, and they all mirror the content. So we would need a multi regional multi provider multi every ISP failure in order to really be unable to access the events that are

203
00:39:03,000 --> 00:39:09,000
notion that one long term, I don't think we've really thought about data retention that much.

204
00:39:09,000 --> 00:39:27,000
We've talked about it as a known eventual problem that once we start struggling with, you know, the ability, our indexes are too big or two spaces out of control, we just haven't hit any of those bottlenecks and our priorities are really to build new, new

205
00:39:27,000 --> 00:39:38,000
solutions at this point, that will be something that we eventually hit a wall on and we'll have to solve for. Yeah, if we if we ever get to the point and like I said that we're really using a small percentage of the kind of the network and the pipeline

206
00:39:38,000 --> 00:39:53,000
we've set up right now so I think we have a long ways to go before we have to worry about that, especially in the spam filtered world because I think as most of you know, you know, 90% of the events that were circulating around.

207
00:39:53,000 --> 00:40:10,000
Nostr, did I say that right? Were spam, you know, so now that most relays are handling those are most of the large relays are handling the spam extremely well these days. To be fair, it's very simple to handle we think that challenge will will get more

208
00:40:10,000 --> 00:40:27,000
challenging. But thankfully because of that, the flow of data is very much under control right now on the network so if we ever do need to do something about data retention, we will make your events available to our, our paid users before we do anything, obviously

209
00:40:27,000 --> 00:40:34,000
we won't just make your events disappear overnight.

210
00:40:34,000 --> 00:40:54,000
All right, I believe Tano you had a question. Yeah, this is from Nostrjack.io from Bill Monroe. The question is, are the current prices sustainable to pay you for your time and infrastructure costs, and will that change significantly if you have a lot more users?

211
00:40:54,000 --> 00:41:12,000
Yeah, that's a good question and we definitely don't have all the answers. And for now our infrastructure costs are certainly controlled, our time is nowhere near paid and we're okay with that we're kind of investing in the future of this network at this, at this time.

212
00:41:12,000 --> 00:41:24,000
But we don't really know exactly how the future is going to look like and because of the very kind of generous time discounts we gave for filter.

213
00:41:24,000 --> 00:41:33,000
We do have a lot of users who will be around for at least two years, so we're going to be supporting those users, and we're going to find a way to make it work.

214
00:41:33,000 --> 00:41:46,000
We might have to raise prices later on, but we're not going to do it for anybody who's paid for time, you know, so anything that we implement will be backwards compatible so to speak.

215
00:41:46,000 --> 00:42:03,000
I would say part of the goal too is that we want to understand what products we can offer that are beneficial for the community as a whole like Welcome is something that's free, and we don't have any plans of that.

216
00:42:03,000 --> 00:42:18,000
That's something that's just beneficial for everybody. And we want to provide value like with this new solution that we're working on where there's there's added value and something that we can, we can use to try to make this more sustainable as well so I think

217
00:42:18,000 --> 00:42:33,000
we're still playing around with with different ideas for how to make it sustainable and we're going to try to, you know, pick the we're going to not try to charge for things that we don't have to and try to figure out how to make it, you know, make it work

218
00:42:33,000 --> 00:42:44,000
long term, but our goal is to be here for a while, and to continue to be able to innovate.

219
00:42:44,000 --> 00:42:49,000
That's good. All right, so I have another one for you guys. I have one from Owen.

220
00:42:49,000 --> 00:42:59,000
He says when you initially set up the really, was it always the plan to turn it on to turn it into a potentially profitable servers, or was it a temporary experiment.

221
00:42:59,000 --> 00:43:14,000
Yeah, so I mean I think I covered it a little bit in the intro but basically I did start with a public relay, and then, and kind of turn to the paid one as a spam filter and like an experiment, and will this be sustainable.

222
00:43:14,000 --> 00:43:29,000
Potentially profitable is certainly the right word, just to be clear, but I think I bought like the first 200 basically users of Nester.wines access anyway, as a kind of a promotional thing.

223
00:43:29,000 --> 00:43:45,000
I do think the paid relay model is makes more sense to me personally as like an as an operator I like to build something that at least is going to cover its own costs and I'd like to be able to as, as I kind of had a thread recently that got some attention

224
00:43:45,000 --> 00:44:10,000
from inside the household, but about me being a bit of a binger when it comes to activities and topics. And, you know, I'm kind of binging this right now. And so, one of the things that you kind of get with Noster.wine is my insane attention on it as well, because that's just basically what I'm what I'm obsessed with.

225
00:44:10,000 --> 00:44:24,000
So, the performance and reliability is is in part what you get from me never looking away from status pages and and seeing every email that comes in within 30 seconds you know so that that that's kind of one of the benefits you get.

226
00:44:24,000 --> 00:44:44,000
And I want to be clear we do, you mentioned we support a few other non Noster projects and Mazin's response times to support is pretty unprecedented even on even when he's not actively pinching so that's something that I think is really special.

227
00:44:44,000 --> 00:45:01,000
I will answer your support requests in a timely fashion that is that is one of my things I don't. If I've received it, it doesn't matter what time it is I'm going to answer it and I can't really focus on anything else until I've answered it.

228
00:45:01,000 --> 00:45:09,000
That's the truth I've experienced this firsthand myself. It's amazing like literally I just ask a question and boom there you are.

229
00:45:09,000 --> 00:45:14,000
Love it.

230
00:45:14,000 --> 00:45:16,000
Oh my god.

231
00:45:16,000 --> 00:45:24,000
I so do you guys would like, like to open it up to live q amp a. Yeah, let's do it. All right.

232
00:45:24,000 --> 00:45:31,000
Audience members, if you'd like you can raise your hand, and this is the time to ask Mazin and Katie.

233
00:45:31,000 --> 00:45:46,000
Yes, don't be shy.

234
00:45:46,000 --> 00:45:51,000
All right, we have CT, come up CT.

235
00:45:51,000 --> 00:45:55,000
Everyone else. Yeah, just raise your hand and we'll bring you up.

236
00:45:55,000 --> 00:46:01,000
Hi, can you hear me.

237
00:46:01,000 --> 00:46:10,000
Yes. Yes. Cool cool what's going on guys. Thank you for having me appreciate I've been a real new to the space and just some learning everything I possibly can.

238
00:46:10,000 --> 00:46:26,000
I just opened up a one paid re, re, relays, and I'm trying to basically do the same thing, cut out spam and make something that's sustainable, you know, so I'm just trying to get your spaces and learn as much as I can I appreciate everything

239
00:46:26,000 --> 00:46:33,000
y'all are doing.

240
00:46:33,000 --> 00:46:46,000
Yeah, appreciate it. We're, we're just kind of out here trying to figure it out ourselves we don't we don't really. It's like one of the things that people always ask me all these absolute questions about the master we don't really have any of the answers either.

241
00:46:46,000 --> 00:46:51,000
So, we're just some guesses and do you have any like specific question.

242
00:46:51,000 --> 00:47:00,000
No, not really. I'm just learning man I just want to introduce myself and say hello I'm going to be around you'll be seeing me. So, I just can't introduce myself mainly.

243
00:47:00,000 --> 00:47:08,000
Nice to meet you. Nice to meet you. Yeah, pleasure to meet you all too. Nice to meet you CT. Nice to meet you CT.

244
00:47:08,000 --> 00:47:12,000
Alright, well yeah, other than that y'all have a great night.

245
00:47:12,000 --> 00:47:15,000
And I'm gonna get back to it.

246
00:47:15,000 --> 00:47:17,000
Have a good night brother.

247
00:47:17,000 --> 00:47:20,000
Good night buddy.

248
00:47:20,000 --> 00:47:25,000
Anybody in the audience have any questions, I'll bring you up right now.

249
00:47:25,000 --> 00:47:31,000
Let me see.

250
00:47:31,000 --> 00:47:35,000
While that's happening, I wanted to ask you guys.

251
00:47:35,000 --> 00:47:40,000
So how did you find Nostra and what's your favorite thing about Nostra.

252
00:47:40,000 --> 00:47:50,000
I found it. Honestly, I don't, I don't know, but probably fiat Jaff something on Twitter probably is where I first heard about it.

253
00:47:50,000 --> 00:48:08,000
And I don't know I'm not a huge social media guy I guess that's not it's not going to be a super popular opinion around here but like it's not I mean I have a Twitter account and I was mostly a lurker I've posted very little, and I, what interested me in

254
00:48:08,000 --> 00:48:21,000
it was basically the protocol and kind of the open nature of it how simple it is and how easy it is to build on and, and I just don't think any of the other solutions are very compelling right now.

255
00:48:21,000 --> 00:48:41,000
So, actually, one of the things that made me snap actually and really moved to Nostra was the. When they re rolled out the for you tab on Twitter, but you weren't at least on iOS you could not save it to go back to your following every time so every time you

256
00:48:41,000 --> 00:48:55,000
went on the app it gave you this like absurd sponsored feed that had nothing to do with the content that I wanted. And that was like that was about it for me. All I really ever wanted from Twitter was to see the content of the people that I was following

257
00:48:55,000 --> 00:49:13,000
in chronological order. And finally, with Nostra, there's a way for all of us to have whatever experience it is that we want. And so that's really what we're going for with filter. The, the follows plus follows filter that we put on is my dream of what a global

258
00:49:13,000 --> 00:49:28,000
should be like, that's not everybody's dream, but it was my dream. So we made it and we use it and we love it so it's, that's, I guess that's really what pushed me over the edge it was it was having somebody else's algorithm that had nothing to do with who I had said I

259
00:49:28,000 --> 00:49:44,000
wanted to see content with shoved in my face. And I would say for me.

260
00:49:44,000 --> 00:50:01,000
I love that. And so, and I've just spent the last week writing front end code I'm just like unbelievably miserable, and I can't design anything that all looks terrible so sorry in advance but it will be, it will be functional because the back end code is good, but that's why we

261
00:50:01,000 --> 00:50:19,000
love Nostra clients because you guys get to do all that's that work. Yeah, and we get to I mean what's really awesome is that we were on we get to play around with it we get to say hey you know what it'd be nice if we had x y z and then we get to just make x y z and so it's.

262
00:50:19,000 --> 00:50:32,000
We should, we should share the, the true origin story of us wanting to build something on Nostra because I kind of forgot to do that because it's a perfect illustration right. We were, we're sitting here with that Nostra that wine paid relay in its early days

263
00:50:32,000 --> 00:50:38,000
and I'm buying subscription, or I'm buying a pay admission fees for everybody right and left.

264
00:50:38,000 --> 00:50:53,000
And I'm like how can we get some content on this network that's like unique. So we came up with the with an idea that I know people like fiat Jeff hate but we were like why don't we just make something that, you know, Instagram creators can just like one click

265
00:50:53,000 --> 00:51:03,000
log into and repost their stuff to Nostra, and then maybe if they get enough engagement on Nostra, sorry, and maybe if they get enough engagement they'll stick around.

266
00:51:03,000 --> 00:51:16,000
So we just like set out to do this we were like this is a fun like afternoon project it should be really easy. We, we like started building it we got like halfway into it and I just started researching like the API docs and stuff for Facebook.

267
00:51:16,000 --> 00:51:25,000
And then I, and then I saw the requirements and it's like, you have to submit videos of your working application and all this stuff.

268
00:51:25,000 --> 00:51:36,000
And you've already spent all the time building it, and maybe they'll approve your access after it's, it's really a ridiculous process that that made us instantly not want to build there.

269
00:51:36,000 --> 00:51:45,000
So, that's, that's really the beauty of Nostra is that if you want to build something you just hop in and build it you don't ask anybody. You just start.

270
00:51:45,000 --> 00:51:59,000
Yeah, I think and we get instant feedback on what it is that's successful so filter was something we wanted. Um, I think it was at the end of that weekend after we had kind of been, you know, working on this product that we threw away that we said oh we can build

271
00:51:59,000 --> 00:52:14,000
this this is really great and we worked on it and we got it out in a couple days and, and it's, you can just build it put it out there and get instant feedback on, you know whether other people want that too and I think that that ability to quickly iterate

272
00:52:14,000 --> 00:52:27,000
and also to just not only decentralized the network but we get to decentralize some of the work that's being done. It's really exciting we get the benefit of everything that everybody else is working on, while we get to work on, you know, components of that too.

273
00:52:27,000 --> 00:52:42,000
So it's, it's energizing. Yeah, we love building stuff for for end users you know, Katie does a lot of building for businesses and her work, and it's a lot less fun, you know, being able to get the instant feedback from users is.

274
00:52:42,000 --> 00:52:57,000
And, and, and like praise when you do something nice. You know most businesses don't really thank you for doing a good job. They just yell at you when you do something wrong so we've really really appreciated the community feedback and it's it's definitely helped drive the

275
00:52:57,000 --> 00:53:03,000
development of filter.

276
00:53:03,000 --> 00:53:11,000
That's incredible man. We had Sergio. Yeah, can you hear me. Yes, we can hear you. Awesome awesome. I am the end user.

277
00:53:11,000 --> 00:53:23,000
I actually, I've never been on Twitter, I signed up for Twitter but I never used it. I never found interest in it. I abandoned social media for like five years found Nostra and I was like, this is amazing.

278
00:53:23,000 --> 00:53:34,000
This is not anything I've ever experienced before. So I'm like, really excited to basically spread the gospel you know like trying to get people in here, and just telling them what it's about.

279
00:53:34,000 --> 00:53:44,000
And I really appreciate all you guys, putting in the hard work as the devs, just the coders, you know I don't have any experience doing that. I'm just a simple club. I'm a bartender by trade.

280
00:53:44,000 --> 00:54:00,000
Like I said I'm just the end user and I think, you know, just everybody accepting me and with with open arms and just like the whole culture and being able to, you know, just take people in and show them what it's about.

281
00:54:00,000 --> 00:54:12,000
It's really powerful and I think you guys are doing a great job and anything that I can do to help, you know, just let me know, you know, I'm here to basically take you back on you guys as coattails.

282
00:54:12,000 --> 00:54:22,000
We appreciate it. I mean we we really wouldn't have been able to spread our product very much if it wasn't for basically a handful of our early users being really loud about it.

283
00:54:22,000 --> 00:54:36,000
So we really appreciate it. Yeah, I mean like I just haven't experienced anything like this before. You know, most of the Bitcoin space is very much one sided where you just listen to things and you're being talked at, basically.

284
00:54:36,000 --> 00:54:51,000
And, you know, with the social media platform you have this interaction where you can actually put your input in and be listened to and kind of and learn from people not just kind of absorb the information that's already out there but you can ask questions that maybe, you know, you can't find the answers to that are

285
00:54:51,000 --> 00:55:02,000
out in the ecosystem at the moment, you know. So I'm always learning things and I really just appreciate you guys for even just letting me, you know, up here and talk to you guys.

286
00:55:02,000 --> 00:55:05,000
Thanks. We appreciate you. Yeah, and if you appreciate you guys.

287
00:55:05,000 --> 00:55:22,000
Feedback like we said we're building things that kind of we identify as things that are frustrating to us but if there's if there's feedback at any point. We'd love to hear it because absolutely was really to make it user driven and not every user is us.

288
00:55:22,000 --> 00:55:31,000
And so we want to make sure it supports, you know, yeah and I'm playing around with all the clients. I'm an Android user so I've really landed on amethyst.

289
00:55:31,000 --> 00:55:40,000
Because I just, I love this interface but I use current Iris clubster, I mess around with all of them, you know.

290
00:55:40,000 --> 00:55:47,000
Yeah, I've pretty much used every client I think I even bought an Android phone so that I could

291
00:55:47,000 --> 00:55:51,000
test our stuff you know I gotta have. Yeah.

292
00:55:51,000 --> 00:55:54,000
Well guys I appreciate it, I have to go to dinner.

293
00:55:54,000 --> 00:55:57,000
But it was good talking to you guys. Have a good night.

294
00:55:57,000 --> 00:55:59,000
Take it easy.

295
00:55:59,000 --> 00:56:02,000
Thank you Sergio. Appreciate it. Thank you Sergio.

296
00:56:02,000 --> 00:56:05,000
Good night buddy. Good night man.

297
00:56:05,000 --> 00:56:10,000
Let me see who else.

298
00:56:10,000 --> 00:56:14,000
We had somebody else but their hand just went down. Anybody else.

299
00:56:14,000 --> 00:56:16,000
Renegade you had your hand up before.

300
00:56:16,000 --> 00:56:23,000
Do you still have a question. He went down to the stage, I mean to the audience I don't I'm pretty sure he.

301
00:56:23,000 --> 00:56:29,000
I don't know do you want to come up Renegade? Just put your hand up if you like.

302
00:56:29,000 --> 00:56:32,000
Let me see.

303
00:56:32,000 --> 00:56:34,000
There you go, he is. Perfect.

304
00:56:34,000 --> 00:56:36,000
Alright Renegade the stage is yours.

305
00:56:36,000 --> 00:56:49,000
Hey man sorry just wanted to pop in quick and I'm just prepping everything because we're coming up here next on stage so I didn't want to take up everybody's time but I was just curious before you guys pop off if you wanted to at all.

306
00:56:49,000 --> 00:57:00,000
I just wanted to do a really quick rundown for somebody who's new to the idea of like Nostra.wine if they've signed up. I know you guys offer a monthly subscription other than the paid relay for like filtering and things like that.

307
00:57:00,000 --> 00:57:15,000
If there was like a really quick breakdown for somebody who has really no idea what the real advantages of that would be and you know if it'd be worth paying a monthly for a filter and what benefits there are of that.

308
00:57:15,000 --> 00:57:19,000
Yeah sure I'll give you the two minute pitch because that's what we have.

309
00:57:19,000 --> 00:57:33,000
Basically, Nostra.wine is just a normal paid relay. It's pretty high performance if I may say so myself but outside of that it's just a normal paid relay and what we mean by that is it's paid to write, not paid to read.

310
00:57:33,000 --> 00:57:46,000
So anybody can read from it publicly but only people who have paid that $18.88 admission fee can write to it. And there's no subscription right now for Nostra.wine we have no plans to add one.

311
00:57:46,000 --> 00:57:56,000
So, once you pay to that one you get write access there. The nice thing about that is it's a very popular relay so you kind of get your content discovered by a lot of people who are reading Nostra.wine.

312
00:57:56,000 --> 00:58:11,000
And it's basically completely spam free because people aren't willing to pay. I've banned exactly one spammer just like repetitive spammer, and it was like in, I think mid February, and there hasn't been a single other person who's paid and spammed like repetitive spam.

313
00:58:11,000 --> 00:58:22,000
So, that's what you kind of get from Nostra.wine. Filter is this kind of totally different product a lot of people are always confused that they're involved somehow but they're completely separate.

314
00:58:22,000 --> 00:58:42,000
The goal with filter is, is to help bridge the gap between private and paid relays so filter aggregates from a bunch of big public relays and you can read the specifics at filter dot Nostra.wine or docs dot Nostra.wine we have kind of two different docs right now because I'm a mess but

315
00:58:42,000 --> 00:59:03,000
there's about 12 relays I think that we aggregate from in broadcast to and that number goes up and down based on reliability basically. And so what you get from filter is you get D duplicated events that are spam filtered, and then you can also add on it, a follows plus follows filter for your global.

316
00:59:03,000 --> 00:59:19,000
So we offer basically optional user configurability for your global feed, so that your global feed for example, if you use the default settings on filter will only show users who either you follow or users that your followers follow.

317
00:59:19,000 --> 00:59:30,000
And we have an extra parameter that you can add to switch that to just the completely unfiltered global besides we get rid of the repetitive spam.

318
00:59:30,000 --> 00:59:42,000
So, what you get basically is one, you get 12 relays in one, that's like the easiest way to think about it. And it's, it's much faster than using 12 public relays.

319
00:59:42,000 --> 01:00:03,000
I would say the benefit to if you've ever experienced turning off all relays other than your paid relays is is really when you start clicking into those threads, you're going to see pretty much all of the details within those threads, even if they're not within your followers the follows net, which is really valuable.

320
01:00:03,000 --> 01:00:16,000
That is incredible. Thank you guys so much for joining us Mazen, Katie. This has been amazing you guys made this hour fly by with all of this incredible knowledge, I'm so grateful to have you guys here.

321
01:00:16,000 --> 01:00:23,000
This has been amazing. I can't wait to guys to have you guys back. And I wish you the best of luck with wine.

322
01:00:23,000 --> 01:00:25,000
You guys are going to do incredible things I just know it.

323
01:00:25,000 --> 01:00:38,000
By the looks of it. I'm going to forever be a client of the. I don't know why I just call myself a client but I'll forever be a plebe that uses your service. Thank you so much.

324
01:00:38,000 --> 01:00:50,000
Thanks. We appreciate it. Thanks for having us. We enjoyed it. Yeah, thank you guys. Love you guys. Peace. See ya. Bye. Thank you. Thank you so much.

325
01:00:50,000 --> 01:00:55,000
Let me actually