From 81bf174c0906ecf8990f6d897a2416f2a509c1c7 Mon Sep 17 00:00:00 2001 From: George Kwabena Appiah <48031636+iamGeorgePro@users.noreply.github.com> Date: Fri, 13 Oct 2023 14:53:06 +0000 Subject: [PATCH 1/5] Rename 01-2023-01-31.md to 001-2023-01-31.md (#333) Signed-off-by: George Kwabena Appiah <48031636+iamGeorgePro@users.noreply.github.com> --- sig-community/meetings/{01-2023-01-31.md => 001-2023-01-31.md} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename sig-community/meetings/{01-2023-01-31.md => 001-2023-01-31.md} (100%) diff --git a/sig-community/meetings/01-2023-01-31.md b/sig-community/meetings/001-2023-01-31.md similarity index 100% rename from sig-community/meetings/01-2023-01-31.md rename to sig-community/meetings/001-2023-01-31.md From 84bfb57daad506b08881a008bfbe05fe7a717d16 Mon Sep 17 00:00:00 2001 From: Drinkwater Date: Tue, 17 Oct 2023 15:19:49 -0400 Subject: [PATCH 2/5] sig-analytics Create 009-2023-10-12.md (#330) * sig-analytics Create 009-2023-10-12.md Signed-off-by: Drinkwater * Update 009-2023-10-12.md Signed-off-by: Drinkwater --------- Signed-off-by: Drinkwater --- sig-analytics/meetings/009-2023-10-12.md | 121 +++++++++++++++++++++++ 1 file changed, 121 insertions(+) create mode 100644 sig-analytics/meetings/009-2023-10-12.md diff --git a/sig-analytics/meetings/009-2023-10-12.md b/sig-analytics/meetings/009-2023-10-12.md new file mode 100644 index 00000000..7b3a6c50 --- /dev/null +++ b/sig-analytics/meetings/009-2023-10-12.md @@ -0,0 +1,121 @@ +# Akash Network - Analytics Special Interest Group (SIG) - Meeting #9 + +### **Agenda** + +- Open discussion +- Cloudmos indexer specs +- Cloudmos Acquisition Proposal: https://www.mintscan.io/akash/proposals/216 (Github discussion: https://github.com/orgs/akash-network/discussions/286) + +### **Meeting Details** + +- Date: Thursday, October 12th, 2023 +- Time: 09:00 AM PT (Pacific Time) +- [Recording](https://v5ixigio4enkf7qhvlmzfptnlx6zkdite2skrznyil2d3xih6ana.arweave.net/r1F0GQ7hGqL-B6rZkr5tXf2VDRMmpKjluEL0Pd0H8Bo) +- [Transcript](#transcript) + +## **Participants** + +- Andrew Averbah +- Anil Murty +- Damir Simpovic +- Denis Lelic +- Jin Lim +- Julius Haus +- Scott Hewitson +- Tyler Wright + +## Notes + +- Tyler recapped last meeting. It was an action item for Jin to connect to Cloudmos Node for indexer. +- Jin talked about getting access to Cloudmos. This happenned successfully! +- Jin is working with https://redash.io/ to visualize data as outlined in PRD. +- Shimpa has a archival node. If anyone needs access, please reach out to him because there are some securirity layers. +- Julius asked if the node is akashnet-2 or akashnet-1. Julius and Shimpa will talk about this online. +- Anil mentioned that there is no way to know what kind of GPUs are on a provider or how many of a certain kind of GPUs on the network. This is an issue. +- There is work on the core roadmap for this. +- Tyler reiterated to look out for announcements for a network upgrade that will include that. + +### Action Items: + +- Jin is going to visualize data. +- Jin will share results with the team in discord. + +# **Transcript** + +_This editable transcript was computer generated and might contain errors. People can also change the text after it was created._ + +Tyler Wright: All right, welcome everybody. To the 9th installment of the Analytics, Special Interest Group. This specialist group deals with all things related to analytics on the Akash network, as of recently. Again, there's been some meetings in the past and discussions about creating an indexer ads of the last couple of months. Cloud. Most has gone through a process of open sourcing and making the indexer available and again, we'll continue to go through the steps. There's also an active PRD inside the city again, analytics Where folks can kind of keep up to date on some of the outstanding items that are being worked on in the background by folks. + +Tyler Wright: Last month at the analytics. they were pitching potential front-end solutions that they can help the Akash network with them, putting a link to + +Tyler Wright: Their website in the chat for folks, they also again got involved in discord and the Monthly basis, I get and continue and work continue Between meetings Jin. I don't want to put you on the spot, but I know that you've been working on some things related to Cloud most index. So I just want to see if there was anything that you want to update the group on today. + +Jin Lim: Yeah, of course. So as you mentioned I was working with Max on connecting to their data solution which went great transformational of our tables are already completely complete and ready for visualization. So we previously considered and I did suggest use of Redash for this effort which is an open source SQL editor and it also Works as the Visualization tool and supports the connection for our current data solution in Postgres. So, currently working with another team on getting that deployed on sh. And then once that's up and running, I can get started on building dashboards to visualize the questions laid out in the PRD. + +Tyler Wright: Perfect. Thank you, Jin. Does anyone have any questions about what Jin's working on next steps? Etc. + +Tyler Wright: Called Jin. Do you know quite yet where you're thinking about posting those visualizations? I know you said on a cash but is there a front end where there'll be available or Where folks in the community can look at things. + +Jin Lim: Yeah, so what I suggest is using redash, so that'll offer the front end side of it as well. Where all the visualizations will be viewable. + +Tyler Wright: Excellent. + +Jin Lim: And then I just sent a link in chat. + +Tyler Wright: Perfect. Thank you very much. + +Tyler Wright: Does anyone else have any questions about again? Anything that Jin, just mentioned the potatoes, to the clubhouse, indexer, and visualizations. In next steps? + +Tyler Wright: I'm going to also attach again the PRD. In the chat to. So folks here have it and again I'll throw it in the notes as well. + +Tyler Wright: quite I just do that. Are there anything else Analytics-wise? Anybody here wants to discuss. any questions about analytics that anybody wants to discuss while there's folks here live that are again actively working on this project. + +Tyler Wright: again. There was some discussions in the past about analytics and indexers, I think there was a couple of organizations that we're thinking about supporting and Neil has mentioned in past meetings. Cloud most has already built a solution that solution is now open source. Thanks to efforts from Overclock Labs, as well as a community pool. And again, that's so data that Jin is already leveraging. And again, we'll make available After he's continued some development. If anybody else has any questions, comments, concerns about anything analytics related. Please feel to feel free to drop them in chat here. Feel free to raise your hand or Free to just drop a message or a question inside this again. Analytics, discord channel. + +00:05:00 + +Tyler Wright: I know that things again are pretty much straightforward right now. I believe there was somebody Zack potentially that was also thinking about building and an archival node. But that might be on hold for right now. So again, I don't want to waste anybody's time. Go ahead Damir. + +Damir Simpovic: Adjust on the archive node, I have spun it up and it's running. So yeah, if anybody needs it, you can use it. Just let me know because it's behind some security features. So yeah, there you go. + +Tyler Wright: Excellent. Julius asked in the chat from Akash, net two, or from a class net one. + +Tyler Wright: Is archive note? + +Damir Simpovic: Archive node is from Block 1. So yeah, all the way Literally from block number one. + +Tyler Wright: Excellent. Go ahead. Julius. + +Julius Haus: yeah, I think there was a network before a passionate 2 and it has also some transactions. because sometimes, I have to query from. Akash Network 1 to get all my transactions. + +Damir Simpovic: are I would you got me. but, I don't know, I'll check but I would imagine that all the transactions are here on this note. But I'll check. + +Damir Simpovic: Yep. + +Tyler Wright: Thank you. Julius. So again there's a damir mentioned, if anybody wants to reach out to access at archival, no feel free to reach out to Damir. Also known as Shimpa on discord and then he can get you access to our time, archival node. + +Tyler Wright: I appreciate that update I'm sure we'll get more updates between The next meeting. Is there anything else analytics why to anybody wants to cover that hasn't been covered thus far? + +Tyler Wright: Go ahead and deal. + +Anil Murty: It's not directly analytics. But essentially one of the pain points we have right now and I think this should be pretty obvious for there is no way to know very quickly. What models of GPUs in general, what resources are available on a given provider or how many forgiven resources are available across different providers. So when you go to any of the analytics dashboards, for example, if you want to just find out how many a 100 GPUs are available right now on the network, there is no way of doing that. And then similarly, when you deploy a application and you go to the least status, there is no way of knowing what exact GPUs you've got. All it tells us to count the GPUs. So there is work on the Core Road map for + +Anil Murty: making these Through various endpoints. So that not just the analytics solution, but also the clients can take advantage of it. So, some piece of information there again, it's not specifically all analytics, but some of it is + +Tyler Wright: Much appreciate that. Anil that is great insight. Again, as a NEO just mentioned, that's very much on the roadmap and it's being worked on by members of Overclock Labs Core team. Right now, please look out for our announcements because again, I believe that there's a chance that there might be a network upgrade that implements some of that prior to potentially this next again, analytics meeting. So. Again, please look out for updates and discord related to that. I mean, thank you very much for bringing that up. + +00:10:00 + +Tyler Wright: So we've talked about Jin's next steps. As it pertains to his work around the cloud, most indexer, which is again now open source and available for folks to use, we talked about Damir, having an archival know that he's running. And again, he and ZJ will, or soon, he J we'll talk offline about question and Julius had. And then Anil just brought up a great point around some additional analytics, or eight pieces of information that will be available as a part of a network upgrade that's coming soon, that'll give folks. And clients more access to the kinds of GPUs that are on the network specifically by count. And then again, what kind of resources are inside the providers. So Out for announcements there soon. + +Tyler Wright: Is there anything else that hasn't been covered that? Anybody wants to cover right now as a pertains to analytics? + +Tyler Wright: Cool if not. Then again, there might be a question that I might ask in the chat over the next couple of weeks at the pertains to How can metrics but again, I'll follow up with that and Just so there's disability for all, but unless there's any other questions and we can certainly wrap up this analytics monthly meeting today. again, there will be additional work that'll be shared by Jin and other members of the team and discord and elsewhere between meetings, the recording and the notes will be available for this meeting very shortly. But does anyone have any questions before? I let you all go for the rest of the day. + +Tyler Wright: if there's no questions again, thank you all for participating. look out for updates from Jin on making. Again, analytics more available to more people and if anybody needs access to archival mode for their own purposes, feel free to reach out to myself or feel free to reach out to shimpa on discord and we can get you access for Again, thank you everybody for joining today, we'll have another analytics monthly meeting in about a month. I believe there's a full schedule next week of meetings, including SIG clients. + +Tyler Wright: Big documentation is coming up in a couple weeks as well as six support. So again, look out for meetings on the schedule, coming up, and again, much appreciated. Everyone's time today. Hope everyone has a great day. Thank you, in for the updates. Thank you. All for participation in the questions. And, yeah, we'll talk online. Have a great day y'all. + +Julius Haus: Fine. + +Meeting ended after 00:13:09 👋 From 52c8d65851c22cf4e5e86bcc1705ab82c27ea8a8 Mon Sep 17 00:00:00 2001 From: George Kwabena Appiah <48031636+iamGeorgePro@users.noreply.github.com> Date: Wed, 18 Oct 2023 02:04:42 +0000 Subject: [PATCH 3/5] Update 006-2023-07-11.md (#331) Signed-off-by: George Kwabena Appiah <48031636+iamGeorgePro@users.noreply.github.com> --- sig-community/meetings/006-2023-07-11.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sig-community/meetings/006-2023-07-11.md b/sig-community/meetings/006-2023-07-11.md index 898230e3..1ffec17e 100644 --- a/sig-community/meetings/006-2023-07-11.md +++ b/sig-community/meetings/006-2023-07-11.md @@ -7,7 +7,7 @@ - Discussion and feedback on the community initiatives within Akash. ## Meeting Details -- Date: Tuesday, June 11, 2023 +- Date: Tuesday, July 11, 2023 - Time: 11:00 AM PT (Pacific Time) - [Recording](https://ldcymjv57vzkxb54husr63qq3saae7qhwos2vcvjoln32ny6cy2q.arweave.net/WMWGJr39cquHvD0lH24Q3IACfgezpaqKqXLbvTceFjU) - [Transcript](#Transcript) From 06513a8ff899dc95bfaf4aac7cdd3ffce7866f6f Mon Sep 17 00:00:00 2001 From: Hewi <121134490+heyhewi@users.noreply.github.com> Date: Tue, 17 Oct 2023 21:12:37 -0500 Subject: [PATCH 4/5] sig-economics: Add meeting 9 and update README (#346) * sig-economics - Create 008-2023-09-06.md Signed-off-by: Hewi <121134490+heyhewi@users.noreply.github.com> * sig-economics Update README.md Signed-off-by: Hewi <121134490+heyhewi@users.noreply.github.com> * sig-economics Update 008-2023-09-06.md updated copy to "Incentive Distribution Pools" Signed-off-by: Drinkwater * Create 009-2023-10-04.md Signed-off-by: Hewi <121134490+heyhewi@users.noreply.github.com> * Update README.md Signed-off-by: Hewi <121134490+heyhewi@users.noreply.github.com> --------- Signed-off-by: Hewi <121134490+heyhewi@users.noreply.github.com> Signed-off-by: Drinkwater Co-authored-by: Drinkwater --- sig-economics/README.md | 1 + sig-economics/meetings/009-2023-10-04.md | 288 +++++++++++++++++++++++ 2 files changed, 289 insertions(+) create mode 100644 sig-economics/meetings/009-2023-10-04.md diff --git a/sig-economics/README.md b/sig-economics/README.md index 7db3abf8..7d3a1018 100644 --- a/sig-economics/README.md +++ b/sig-economics/README.md @@ -20,6 +20,7 @@ Meetings happen every [First Wednesday of the Month](https://calendar.google.com | #6 | Wednesday, July 5, 2023 10:00 AM PT (Pacific Time) | [Link](meetings/006-2023-07-05.md) | [Link](meetings/006-2023-07-05.md#transcript) | [Link](https://h2cpwx7cvuhq7gn2rqgbm2nn34r3ie7vppfmgnbpyzmp3sbnqw3q.arweave.net/PoT7X-KtDw-ZuowMFmmt3yO0E_V7ysM0L8ZY_cgthbc) | #7 | Wednesday, August 2, 2023 10:00 AM PT (Pacific Time) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/007-2023-08-02.md) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/007-2023-08-02.md#transcript) | [Link](https://fc2diejm5vhix53dnrxh7dh2bflkuufterrrnmjb54ng4ojumtkq.arweave.net/KLQ0ESztTov3Y2xuf4z6CVaqULMkYxaxIe8abjk0ZNU) | #8 | Wednesday, September 6, 2023 10:00 AM PT (Pacific Time) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/008-2023-09-06.md) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/008-2023-09-06.md#transcript) | [Link](https://nlzzmaec3zpmorkyh3f4w3tep2hy444funhf645cqroo7x4qtusa.arweave.net/avOWAILeXsdFWD7Ly25kfo-Oc4WjTl9zooRc79-QnSQ) +| #9 | Wednesday, October 4, 2023 10:00 AM PT (Pacific Time) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/009-2023-10-04.md) | [Link](https://github.com/akash-network/community/blob/main/sig-economics/meetings/009-2023-10-04.md#transcript) | [Link](https://zq6rzise7b65e5o4uq2t2dgfvdu5fjj4p2aluihozhraqk7eyyca.arweave.net/zD0cokT4fdJ13KQ1PQzFqOnSpTx-gLog7sniCCvkxgQ) ## Leadership diff --git a/sig-economics/meetings/009-2023-10-04.md b/sig-economics/meetings/009-2023-10-04.md new file mode 100644 index 00000000..611e0df4 --- /dev/null +++ b/sig-economics/meetings/009-2023-10-04.md @@ -0,0 +1,288 @@ +--- +title: "Economics SIG Meeting #9" +key: "sig-economics-meeting-9" +date: 2023-10-04T18:00:00Z +contributors: + + +- Andrew Gnatyuk +- Cheng Wang +- George Pro +- Jia Yi Tan +- Joao Luna +- Robert Koschig +- Scott Hewitson +- Tyler Wright +- Zach Horn + + +--- + +# Akash Network - Economics Special Interest Group (SIG) - Meeting #9 + +## Agenda + +* Updates since last monthly sig-economics meeting. +* Open up to community for ideas or concerns + +## Meeting Details + +- Date: Wednesday, October 4, 2023 +- Time: 10:00 AM PT (Pacific Time) +- [Recording](https://zq6rzise7b65e5o4uq2t2dgfvdu5fjj4p2aluihozhraqk7eyyca.arweave.net/zD0cokT4fdJ13KQ1PQzFqOnSpTx-gLog7sniCCvkxgQ) +- [Transcript](#transcript) + +## Participants + +- Andrew Gnatyuk +- Cheng Wang +- George Pro +- Jia Yi Tan +- Joao Luna +- Robert Koschig +- Scott Hewitson +- Tyler Wright +- Zach Horn + +## Notes + +**Introduction and Updates** +- Scott Hewitson initiated the meeting, inviting attendees to discuss various agenda items. + +**Discussion on IDP and Community Pool** +- Cheng Wang differentiated between IDP and the community pool, explaining that the latter serves as a general repository for funds. +- The IDP aims for clearer allocations and accounting of funds, though Cheng expressed it's not a primary concern at this time. + +**Token Economics and Infrastructure** +- Cheng shed light on the other "Akt 2.0" changes, noting that shared security and chain-related developments could need addressing before introducing IDPs. +- Cheng underlined the need for understanding engineering bandwidth and priorities before rolling out pools like the IDP. + +**Bounties and Community Initiatives** +- Cheng discussed the topic of bounties, indicating that as more specifics about bounties get ironed out, the structure for them becomes clearer. +- The talk alluded to the possibility of adjusting the structure as they get more insight into how bounties function and their impact. + +**Community Questions** +- **Question 1**: Andrew Gnatyuk suggested creating a mechanism or UI to track how funds are added to the community pool. +  - **Answer**: Cheng recalled past dialogues on this matter, acknowledging the significance of this transparency. He assured to revisit the topic with the core team. +- **Question 2**: Andrew emphasized the need to understand and show the contributions to the community pool, e.g., take fees vs. inflation. + - **Answer**: Cheng concurred, reiterating the importance of this clarity before launching IDPs. + +**Closing Remarks** +- Scott asked for any final thoughts and promoted Discord as a platform for continued discussions. The meeting concluded with gratitude expressed by Cheng, Andrew, and others. + + +### Action Items + +All are welcome to contribute! Anyone who wants to contribute please visit sig-economics in the [Discord Server](https://discord.gg/akash) + +## Transcript + +This editable transcript was computer generated and might contain errors. People can also change the text after it was created. + +Scott Hewitson: All right, good morning everybody. This is the eighth or no 9th installment of Cig economics. Where we talk about all things about the token, Akt incentives, and long-term ways to promote the network using them. last month, We talked about a few things from Proposals, that will pass. A little bit about idps. Again, the incentive distribution pools, a little bit of downies and fielded some questions about take rates stats and things like that. + +Scott Hewitson: I don't have a major outline for today. So when that kind of the case, we kind of just open things up or, I can let Chength get started and see if he wants to add anything. And then we can open it up to community members and see if anyone has anything they're working on or questions or concerns things they want to address. So, over to you Cheng. + +Cheng Wang: All right. Thanks y. Hot potato received. I do have a little bit of a list to go down, thank you so much. Thanks everyone again for joining these things. It's always great to see community members familiar faces and hopefully some new faces over time. I know, economics isn't necessarily the + +Cheng Wang: Sexiest thing and people typically don't want to make dramatic decisions or just decisions in general, that impact economics, because those decisions tend to be contentious. But I do want to say for the record that everyone if you have a point or a thing you want to discuss, Please do chime in openly as At the beginning of this call, there are community members who have questions and do have ideas and thoughts that they want to address. So, please do let it be, state in publicly, if you feel comfortable, best ideas will ultimately emerging, we'll have our chance to just kind of shake it out. It's one of the reasons why all of our proposals and thoughts with regards. The economics goes on chain and goes before that even to Github discussions and also covered in this particular medium as well. + +Cheng Wang: I'm going into a categorical thing that we really started doing earlier this year and will continue to really flesh out and hone over the next couple of months, through the end of the year. And of course, start strong in 2024. And that particular category is bounties. I know, that kind of started with the Red Test net. If you can think about it is really an aggregation of bounties for doing specific tasks and things like that. And we have been working in codifying a couple of those things for contributions in terms of content and little things are being developed like sdxl that's live on a college but those are really bespoke + +Cheng Wang: And one-to-one, I would love to, engage in more. And if people have interests or thoughts, as far as how they can contribute for development purposes, maintaining awesome ai stuff doing, particular contributions. Now that's really top of mind for us. Right Is surround Aiml stuff like what ZJ pushed out for sdxl, that stuff is absolutely incredible. Just to kind of be able to see and be able to formulate bounties around that too. There are anyone here interested in that vein? I know those require quite a level of technical expertise and understanding of the subject matter. But if, someone or yourself or interested in Please do let us know, hit us up. If you have an interesting idea, please don't be shy. + +Cheng Wang: offer it, we always are looking for new ideas to kind of tinker around with and hopefully it sticks. In addition to that bounties for content, right? Whether it's written video, all that stuff, I know logistically, there's a lot of coordination costs. And, just coordination cost to ensure things online brands. All that kind of stuff, designs, etc. But we're internally starting to think the core team is rather around, sourcing content from the community via bounties as well. And, ensuring that there's high quality and all that stuff similar to how kind of the Web page, right? The website excuse me has evolved So that's along the lines of bounties. I'll pause there. Any questions or thoughts on boundaries? + +Cheng Wang: Always Andrew far away. + +Andrew Gnatyuk: Yeah, hi guys. So all right, + +Cheng Wang: Man, how are you? Love it. + +Andrew Gnatyuk: Yeah, I'm fine. So up, just one is a suggestion. Did you want to put for example a separate page in our site with the bounties or on our discord just for people to like so we could just either tell people Hey there was a link if you find yourself useful and there is a bunch of bounties with rewards if you want to just check it. Obviously you will need a generator there, some tech guy for the technical bounty. some for the content creation, bounties like moderations. But still, I think it's gonna be useful or we can use some additional sites with whereas lots of bounty hunters and they could just create the content creation. there's the stuff with the sides with already that already has a bounty hunters. It's kind of came back fire because there are guys, doing some cheap content and what's the payment? that's like the bad stuff. + +00:05:00 + +Cheng Wang: Yeah. + +Andrew Gnatyuk: So probably + +Andrew Gnatyuk: The additional page on our site with the bounties could work for us. Better maybe not much of the content and AI bounties but still they will be more quality or something like that. Just a suggestion. + +Cheng Wang: Great suggestion and your brain has gone through I guess the iteration cycles. our brains ourselves have gone through. It's exactly the same pathway as + +Cheng Wang: first is And before we just go gangbusters and put out a bounty page and hope for the best, we really need to manually test some of these things, Just to go through the process, see the content itself, how it appeals the people, How do we get the content? Whatever. The thing is that, get the thing done, Get the thing, iterate through it, get the thing approved, if it's, good to go. And if it's not, how do we push it back and give feedback all that kind of stuff, And coordination there, of course, is the kind of overarching umbrella for this? And so, a lot of this stuff we've been doing right now, almost kind of beneath the waves if you will just really testing some of these out on a one-off basis or on a few off basis, to really understand how that goes. And then for things that make sense to + +Cheng Wang: Be pushed out and more structured public bounties. I completely agree with you. That's something that I think is probably end of the year or maybe even a 2024 items, I have a dedicated page or something like that, that's accessible to everyone. I agree with you there that we don't want to necessarily go to bounty Hunter sites, maybe sooner. Thai publicly Tyler Wright, for the record said it may happen,… + +Tyler Wright: it's + +Cheng Wang: maybe sooner. So it is now forever part of the Internet. But yes, the + +Cheng Wang: Bounty Hunter sites. I think, definitely not. Because we want to curate for quality not quantity necessarily Number one, and number two is the more crap comes in, the more low that puts on us. And potentially they go see how core team, as well as the rest of the community as well to Curate and go through some of the stuff, So we definitely want to start slow and start really hit our core audience before we kind of go out with it, and furthermore going starting small will allow us to further librate. These bounties maybe at some point in time. They can be mass, pushed out and on mass, So if they're well structured and there's, well known and Cohesive Way where all the checkpoints and all that stuff and curation can be done relatively seamlessly and without much manual intervention but I think well taking on all those points. + +Cheng Wang: Go for here. + +Scott Hewitson: Yeah. Yeah, one thing I just wanted to kind of back things up a little bit on bounty. So the bounties were kind of talking about it and have been iterating on come from the Community Support governance proposal. So the first one was in the one for Q4 just passed a few days ago. And So part of the community support was, vanguards and then bounties and note-taking. There hasn't been really anyone doing notes. So that's why you can see any spend on that, but the bounties. change the saying, I don't know. Yeah Cheng was saying it's been a little bit more one-off one to one it's kind of like a little bit of a pilot so it's five grand a month and downies and we've been + +Scott Hewitson: Haven't been able to spend all of it but I do like Andrew's idea of maybe not necessarily a page but if we could do something in discord or whatever, go into the insider it's like, here's the open bounties for insiders. to get rid of bounty hunters who people are just trying to, they're not aligned right? So whether it's kind of in vanguards insiders whoever,… + +Cheng Wang: Yeah. + +Scott Hewitson: so if you're not an inside or want to be insider, hit up Adam Wozni for sure because that might be a good avenue for that. What's up, Andrew? + +00:10:00 + +Andrew Gnatyuk: Yeah, there's actually a great idea. as the first steps we can just create the closed channel on our discord. Just to insiders and user will be not enough insiders to submit the Bounties. We can open it to the validators and providers just to see how they are involved. And after that, we just can open that for the community So there was just separate channel where we just publishing all the bounties and people just submit in them to whom ever owns the channel or something like that. + +Scott Hewitson: Yeah I mean I like the discord drought. I don't know if that's feasible or if there's some blockers to that or thoughts on Ty, I don't know. What do you think? + +Tyler Wright: No, I'm a fan of the dedicated channel idea. I know we also have inside our vanguards channel that we can theoretically pin a spreadsheet to, if we just want to keep conversation in there. But this is something that I think we've been talking about as Overclock Labs team is again, trying it out with folks and insiders in vanguards and building relationships with folks, that we know had certain abilities. And then continuing to branch out from there to different members of the community and beyond. So, I think we can talk and Andrew, I might ping you again after this meeting, just to get your thoughts. But I don't want to do is have so many channels that people are just fragmented, I think is happening in the past, But we could figure out the best route. So that + +Tyler Wright: I get those that want to participate in the donkey program or want to earn some extra Akt by supporting things on the network is easily accessible for them. We've talked about hosting in a github in the community sections as well as in discord again, but we'll continue to find to the process of the next couple days. + +Scott Hewitson: George what you got? + +Scott Hewitson: your hands raised. + +George Pro: And thank you very much. And I wanted to say that concerning the note taking and I'll be working on some of the news. So in case there's any new that I have to work on You can let Tyler know then Tyler, will I? Let me see that. I work on it as soon as possible. + +Scott Hewitson: . + +Cheng Wang: Beautiful. + +Tyler Wright: Yeah, just to add to that. Thank you, I know that George has stepped up and is going to be more hands-on with some notes on the various 6 and working groups again. There is no taking Akt available. So I've been taking notes for a lot of the CIGS and working groups for free is, part of responsibilities and growing the network again. But there is Akt available for folks that want to help take some notes for very sigs and working groups. So again, shout out George. Go ahead Andrew. + +Andrew Gnatyuk: Yeah, we actually can fair you die. We have some funds from the community pool self. If you need semi criti, just reach out to heavy, + +Tyler Wright: Yeah, yeah exactly. And I don't think we'll do that … + +Scott Hewitson: Is defeated. + +Tyler Wright: but I appreciate you thinking about me Andrew. + +Scott Hewitson: Hit this man paid. + +Scott Hewitson: Andrew, did you want to say something else? + +Cheng Wang: A nice. + +Tyler Wright: The other thing,… + +Andrew Gnatyuk: Yeah, no. + +Tyler Wright: I'll say real quick. No, Andrew is based on this conversation. I know you work with a number of folks closely on the insiders and vanguards, it'd be great if you can again continue to spread the word about this program. I know that there have been folks in the community already that have been participating from the bounties and helping bring, workloads online or working on editing projects at the part of the inside of vanguards. But if there's anybody else that is technical that would benefit from some bounties, just like ping with an insiders and vanguards, we're trying to let people know but Andrew, your health is always helpful. + +Cheng Wang: Or Yeah, I want to emphasize once again that if we don't have a bounty for something that doesn't mean that particular item or subject Off the table. because we had an economics to it or we had any thought of it. So if there's something that you find might be worth while and you want to throw up for consideration or discussion, please do so. + +Cheng Wang: Enter. + +Scott Hewitson: Go for it. + +Andrew Gnatyuk: Yeah, and just a quick question, and for example, if I'm putting this on discussion, how are you doing? Just Github discussions for certain bounty, to create rules rewards, etc, How should we do that? + +Cheng Wang: Yeah, for I think there is ways to do it. It depends a little bit of a out answer. But it depends and specifically here depending on how Big or well thought out. It is hey taking you repost this thing. Let me just like you repost, it just in case people who missed it. Here. So, for example, this particular thing is + +00:15:00 + +Cheng Wang: Very well thought out and written out, I'm not as technical so I can't truly assess the technical feasibility of this one way or the other, but something like that would be clearly on its surface, very useful for the network, And is quite technical and is pretty well thought out. So if you have just an idea of what something ought to be, I think if it's just, hey, I have an idea and it's something I haven't really Thought, I just wanted to block throw it out there. I would say, discord might be a better place to engage in more active discussion so that you can refine that idea within the insiders or in the community in general, if you wish. But yeah, among insiders in vanguards might be a good place to start that and then once that idea, hope maybe if it gets a little bit of traction and you refine it a little bit further than posting it to Github discussions. + +Cheng Wang: If it gets to that point where you're like, Hey, this is well fleshed out. I'm gonna write Or we are going to write this whomever it might be to get funding for it. That would be a good place for it. So I would say that would be the evolution something from a kernel of an idea to something in discussion and going through iteration and for final kind of preview and final discussion within Github itself + +Andrew Gnatyuk: Got it, thanks. + +Cheng Wang: Yes. Hi. New Jersey. + +Cheng Wang: Said exactly, indeed. + +Cheng Wang: All right, and the other items are, of course, events and hackathons which are well, formulated and thought out. And those are very well structured. Thanks to and Huey was in here. He's out and Cosmo verse rep in a cautious yell. But yeah, there's that side of things as well, but, don't need to + +Cheng Wang: Beat that drum a little bit, veto drum, too much. The only thing I'll say there is for hackathons. Same thing, if you yourself want to participate and know, someone who might want to contribute there, relevant HACKATHON, participants for specific things that the network needs is always going to be welcome. So definitely take a look at those as much as you can. A lot of insiders have participated in this most recent one and similar to the vein of what you're referring to as far as bounty hunters Initially right before the submissions towards the tail end, we're very much cost specific in the beginning. It was just people who just develop things and just threw them into every single category and every single Packathon just to see if they can get some free money out of it, spray and ray. And so those people are gonna exist everywhere and that's just kind of how it is. But + +Cheng Wang: The extent where we can galvanize community or encourage new participants to come in and contribute in a meaningful way. Let's do it. there are two more items that I want to talk about really quickly. And then we can turn it over the business of I'll start with a smaller thing recap for the year. So as we draw, I know to the last quarter of the year and the last month, so here to the holidays and such a special in the US. I think really kind of goes by pretty quickly. I would say that he was gonna leave the charge here for a recap of all of the kind of bounties incentives. The events, like all the stuff we've done kind of a look, back on the year because I think, as a network, we've really done an incredible job with the community and, the team here putting together, I think a vehicle mechanism that really + +Cheng Wang: I think allows a community to contribute and it's super meaningful way and it's incredible to see the progress that we made thus far and the speed at which we're able to move. I know, in some cases a little bit slower than others but it's been awesome and we want to continue to grow that. So we're gonna do a kind of a year in review type of thing, closer towards the end of the year. and and in terms of feedback or comments, I love to get that thing posted and know how we can do better or if there are community Thoughts around there. Please share with us. + +00:20:00 + +Cheng Wang: And the last item that I have is a big one, provider incentives and tenant incentives so that's something that was discussed early on in Akt 2.0. And now that GPUs have been online for a couple of weeks now, over a month that I guess about a month coming up right at this point in time and things are stabilizing. We're starting to really I get her legs under us. So we're internally starting to think about structuring the initial kind of pilot set of providers incentives, and tenants incentives. So there's a lot of work or just thought, being put behind it. nothing refined yet. That is shareable to the community, but just want to make sure that everyone is aware. That this is something the internal core team is thinking about, There's a really important because, again, it's Greg said, Multiple times publicly, Is a caution network is pretty incentives. Unlike the final coins, etc. + +Cheng Wang: Where you bootstrap the tanas apply and then usage dramatically lower than supply. We're at a point where kind of teetering and in a bit of a balance, I know there are a lot of GPUs approximately like 40 or so percent that are underutilized but that's simply kind of based on There are certain models that are more desirable than others and of course we're trying to push right as a community as a team demand. At the same time, we want to grow supply in the areas where the demand needs to be or wants, right? So we can meet the man where it is. So in that vein lot of potential interviews happening an opportunity but interviews happening with users who want to use a caution who want to learn more about it, we want to learn essentially what their use cases are and things like that and what they're kind of pain points are so we can address them. And provider and tenants incentives are going to be really important to really essentially get them on boarded and get them using akash so that we can get additional feedback and + +Cheng Wang: Of the console, the network, etc. All that stuff and the incentives are a huge part of that because as you experiment and as you try new things, it's kind of like credits, So we want to make sure that that Benefit is there for people in the community. We're just users who will want to come in and really try something out. Seriously. And so, Definitely keep an eye out for that. That's something I think in the initial draft or something should be published within a few weeks is my guess. Hopefully, two weeks or so publicly. And we'll get to work on that. I know the first kind of tip of that iceberg is you can kind of interpret this way was the thumper item that was posted by a Neil and get out discussions. So that's a one tiny small, use case around one particular. + +Cheng Wang: Group or I guess project that wants to do something on a caution network like a foundational model training But as that kind of grows, right, we'll definitely need something a bit more robust and cod So keep an eye out for that. That's Is there any questions around that item? Please say so otherwise. + +Scott Hewitson: All right, so Cheng if you don't have anything else I wanted to give a chance for anybody on the call to bring up any ideas, thoughts, concerns or just tell us how you're feeling. So, open it up. + +Scott Hewitson: All there we go. Thanks Andrew. You're always there to help with the question. + +Cheng Wang: Love it,… + +Cheng Wang: Andrew's best. + +Scott Hewitson: Appreciate you. + +Andrew Gnatyuk: Yeah, I just wanted general questions so with the economics 2.0 and what are with our daycare maker fees stuff like where we had that + +Cheng Wang: Taker maker. So, the first iteration of take fees is only take fees, right? There's no maker fee yet because implementing what we realize is. In the original discussion, the idea taker maker fees makes sense, That The tenant coming in, should pay a portion of it and the providers should pay a portion of it taker maker fees here. We've kind of cobbled it all together in the initial iteration into a single take fee, kind of at the end of that flow of fun flow because the other side of it, the implementation of which is really, really challenging for a couple of reasons. it just technically more challenging because you need to put it into two places. + +00:25:00 + +Cheng Wang: Two is Where do you like Here's a kind of pulling back the curtain a little bit. When we're talking about some of these things is Where do you do the calculation? right, let's say it's a long-running lease and the payment and Akt being of all little asset is As your tenant coming in you initially strike the lease and the least, akt's at, 90 cents. And by the time the Block Prince over the course of time, or as the lease runs over the course of time, Akt fluctuates up or down the take percentage, do you put it into Astro, where do you put it, How frequently and what price do you snap it at? And so that's kind of a little I guess you could just some percentage of the units that are at hand here that might make it easier. so those are some of the things that are a little bit. + +Cheng Wang: That we're challenging. It's like the methodology with what, methodology will be apply. that we take an oracle feed from osmosis will we take it from elsewhere to calculate price? Do we take it? How frequently is that taken? And where that would be put? And in addition to that putting it somewhere else, you just putting in another place. In addition to that, how I'd say, there's another challenge, where, if you're 10 and you come in you said, you establish at least, let's say the provider goes down or whatever it is, And it stops, How do you refund that? It has to basically go into another escrow account, If you're a tenant and your lease goes down for whatever reason, your workload goes down. So, there are a lot of, logistical challenges of actually implementing the true Maker and taker, + +Cheng Wang: Yeah, the true maker and taker style take fee, if you will, that was envision, akc2 auto. So this current implementation, makes it a lot simpler. The idea here is, if you're a provider and you charge, let's say a hundred dollars for something, And you're gonna lose 20% if you're charging USCC on the take. Ideally here, you would be calling some of that you would raise your price to a point where, you cover part of that fee, and you pass part of that fee, onto the tenant. So it should ultimately balance out and there's no existing discussion right now to Implement the item like I was discussed in an ak2 autos simply because of complexity and there's a lot more important things that we have to knock out as a network. Hopefully, that answers your question. + +Andrew Gnatyuk: Yeah. Yeah, it does mostly things, but you have at least estimate, dates, like monsters, like half of the year and Okay. + +Cheng Wang: I don't know because it hasn't been discussed. so what I want to assuages that, the take fee and make fee have been basically merged together in the current implementation into just one. Fee. Right? And so that's going to be the implementation currently, as far as, is it going to evolve over time? I pretty sure these things will, but ultimately yeah, I can't say for certain because if you look at typical fees If you look at things that we all use on the Internet, whether it's a food, delivery service or Netflix, or whatever that whatever, you have an iPhone or Google phone, whatever it is and you buy an app and you purchase that, Apple or Google, Uber, whatever, third Party delivery app, they take their cut right from the merchant. + +Cheng Wang: And the merchant ultimately bills you higher I guess at least in in my experience when you order food from a restaurant, if you were to physically, go and get takeaway, it's $10, right? But if you get it on an app, it's like 13 dollars because The restaurant knows that there's a fee taken from their revenues, so they have to charge the end consumer. So in this way it's, tried and true very similar to that type of methodology. So as far as I changing it or further enhancing it, or making it more fine grain I do there hasn't any thought around it. So I don't want to, put my foot in my mouth, but that's just kind of where things stand right now. + +Andrew Gnatyuk: The other things. + +Scott Hewitson: All right. Luna, I know you just joined in the middle of us. at this point. We've kind of opened it up to community and seeing if anyone has any questions, comments, concerns. So I'll open it up to you if you want to say anything. And then if not I think we can adjourn. Alrighty. Andrew. Hit it. + +00:30:00 + +Cheng Wang: we're not. No so those are great question and those are totally separate items. So the IDP thing has been sitting out there for It's as dusty as they come It's a get-up discussion. That's gotten no love because it's rather boring, Just like creating some accounts and things like that. So that is definitely something that needs to. I think I can revive it a little bit and just see what people say, but the IDP accounts are + +Cheng Wang: In iteration of the community pool. the community pool is basically a one single giant thing where everything goes right, and then we write individualized proposals for, Maybe this is a particularly use case. We would like to extract funding for and Maybe there's that. And so it's a catch, all the IDP attempts to make that a little bit clearer with the community pool being up component to it, but not everything else, So that there's a clear delineation where things are going and so as far as working out creating some accounts there though it is. + +Cheng Wang: I think I would categorize it as a nice to have, for the time being and not necessarily an essential component because the current workflow, we have of Github discussion into governance. Proposal is quite good. I think, and it's function, quite well, for the time being, and you don't want to over engineer up front, So just to create more structure and red tape, because then there's a debate of, this is going to that pool or that pool or this, or the other. So ask things kind of like, bounties, As we start to manually flesh things out. The structure becomes clear something And we should definitely implement, but I have to say the precursor to that is a lot of the Akt 2.0 stuff. + +Cheng Wang: Right, to take fees, we implemented, That's great. Another thing like the chain kind of things have to be implemented as well. the shared security stuff. I don't know where that kind of currently resides because that's a component of it, mesh, shared, security, all that kind of stuff has to be. I think implemented in order for IDP stuff to make sense or else right now. it's just inflation rewards essentially, separating into these particular pools which doesn't really make a lot of sense currently. So there's a lot of pre-work for the economics of the token itself that have to be done prior to these pools making sense or those two things have to be done. Concurrently + +Andrew Gnatyuk: Got it. And what last question or more suggestion is zero. Is it possible to create? I don't know, separate UI or separate place on the accusatics to see for example, one when we Implements Take fees, Let's full form, for example, our community pool has been grown for us some portion of a kitty Wright but no one knows. Why does it comes from before that, we know that it was just from change commissions, from the inflation stuff. when we have the feast, it would be great to see … + +Cheng Wang: Up. + +Andrew Gnatyuk: Hey for the past 24 hours community pool, Has been funded for four or five hundred Akt, and 400 was from take fees and 100 posts from the inflation or just change fees, something like that. just to see how this stuff works. + +Cheng Wang: Yeah, I think that was kind of talked about a little bit in passing somewhere. I want to say in one of these segregor, working Group meetings. analytics. Okay, probably. Yeah, I think this is a visibility that we would love to know just in terms of how much our take fees actually coming in and stuff like that, Because that will ultimately be. So that's another great call out Andrew, It's a prerequisite for idps, In order to know the accounting of, what is contributing to the pool and take fee stuff. Should go into a take feed bucket, Not just like broad community bucket, right in the + +Cheng Wang: Version of IDP that was written some months ago, So, that's another great example to your point though. I'll chase on that a little bit with the internal team to see where exactly that was covered by your 100% right. It definitely should be something put on the list. As far as things, we need to outline. + +00:35:00 + +Cheng Wang: Be able to report on in and understand accounting. I agree fully. + +Scott Hewitson: Alrighty. If anybody else has any other questions, please bring them up now or forever. Hold your peace for one month I guess or… + +Cheng Wang: Or hop into discord. + +Scott Hewitson: in discord's a little slow for sig economics but Honestly, if anyone posts in there, I'm gonna respond, I promise. Yeah, so with that, we can close today's meeting, appreciate everybody being on the call, appreciate chain, giving us some updates, appreciate everyone asking questions. And Making everybody think a little bit. I appreciate all that stuff. So with that I'm going to end the recording and we'll see y'all next week or in discord. + +Andrew Gnatyuk: Yeah, thanks guys. + +Cheng Wang: Thanks everyone. + +Andrew Gnatyuk: You bye. + +Cheng Wang: Thank you Scott. Thanks everyone. + +Robert Koschig: Yes. + +Meeting ended after 00:36:19 From 0a2c284b6bbe35cc8104f33fe6715c9069b12abf Mon Sep 17 00:00:00 2001 From: Drinkwater Date: Tue, 17 Oct 2023 22:16:43 -0400 Subject: [PATCH 5/5] sig-analytics Update 009-2023-10-12.md Signed-off-by: Drinkwater --- sig-analytics/meetings/009-2023-10-12.md | 39 ++++++++++++++++++------ 1 file changed, 29 insertions(+), 10 deletions(-) diff --git a/sig-analytics/meetings/009-2023-10-12.md b/sig-analytics/meetings/009-2023-10-12.md index 7b3a6c50..3f5ba6fd 100644 --- a/sig-analytics/meetings/009-2023-10-12.md +++ b/sig-analytics/meetings/009-2023-10-12.md @@ -24,16 +24,35 @@ - Scott Hewitson - Tyler Wright -## Notes - -- Tyler recapped last meeting. It was an action item for Jin to connect to Cloudmos Node for indexer. -- Jin talked about getting access to Cloudmos. This happenned successfully! -- Jin is working with https://redash.io/ to visualize data as outlined in PRD. -- Shimpa has a archival node. If anyone needs access, please reach out to him because there are some securirity layers. -- Julius asked if the node is akashnet-2 or akashnet-1. Julius and Shimpa will talk about this online. -- Anil mentioned that there is no way to know what kind of GPUs are on a provider or how many of a certain kind of GPUs on the network. This is an issue. -- There is work on the core roadmap for this. -- Tyler reiterated to look out for announcements for a network upgrade that will include that. + +## Meeting Notes + +### Review of Past Meetings and Indexer Updates +- Tyler mentioned past meetings and the open-sourcing of the indexer by Cloudmos. +- Tyler referred to an [active PRD](https://github.com/akash-network/community/blob/main/sig-analytics/prd.md) for analytics updates. +- Discussion of the previous month's presentation of potential front-end solutions by the analytics group. + +### Progress Update on Cloudmos Indexer and analytics Dashboard +- Jin Lim provided an update on working with Max to connect to Cloudmos data solution. +- Jin mentioned the completion of data transformation and readiness for visualization. +- Tyler asked about the location of the visualizations, to which Jin suggested using Redash as the front end. +- Jin discussed deployment plans and building dashboards based on the PRD. + +### Discussions on Archival Node +- Zack was mentioned as potentially building an archival node, which may be on hold. +- Damir Simpovic reported that he has spun up the archival node and it's running, offering it to anyone who needs it. +- Julius asked about which network the archive node is from (Akashnet-1 or Akashnet-2). +- Julius mentioned needing transactions from Akash Network 1. +- Damir assured that he would check and suggested that all transactions are on this node. + +### Other Matters discussed +- Anil Murty discussed the need for quickly identifying available GPU resources on the network, both in terms of total counts and provider-specific information. +- Anil mentioned that work is underway to provide this information through various endpoints to benefit analytics solutions and clients. +- Tyler expressed appreciation for Anil's insights and mentioned that the roadmap for this feature is being worked on by Overclock Labs Core team. Suggested keeping an eye out for potential network upgrades. + +## Action Items +- Interested participants to reach out to Damir (Shimpa on Discord) for access to the archival node. +- Jin to continue working and building dashboard based on the analytics PRD. ### Action Items: