Discuss and review: community rewards process + cycle from 3/26 - 4/29 #44
Replies: 9 comments 37 replies
-
|
First of all, we have to make this process dead-simple. For example, AlgoRhythm started with adding comments at the end of each issue with the recipient and related rewards. His reasoning was that it's just a simple copy/paste. But, and that's from the real-world experience here in ASA Stats, for example @MGHQ-YT1 wouldn't have done that step even if his life depends on that. And so we'd end without that information at the end of the issue (better outcome) or some other members would complain about that aloud in public space (perfect outcome). But, the thing is that they wouldn't have complained aload, they just feel awful why such a task is valued the same like their that's much better. And then such people simply stop contributing. So, we need this to be organized and conducted in a bare minimum number of steps and the rest should always be deleted to not make invalid precedence. So, a member of the community creates a thread entitled just like this one on Saturday after a cycle ends. After that, the same member fills the spreadsheet like AlgoRhythm has explained above and after that he/she copies and pastes the spreadsheet values in the thread. The thread is open for the next few hours (or days in this case) until all the contributions and related levels are defined. The community then checks for eventual missing contributions and defines the levels for each contribution. Finally, the list for allocation is created and @ipaleka (for now...) is pinged about. |
Beta Was this translation helpful? Give feedback.
-
|
The thing is that maybe copy/paste of that table here is not needed, we can just simply link the spreadsheet and point to the name of the sheet and the name of the section. |
Beta Was this translation helpful? Give feedback.
-
|
I agree - just as easy to click the link into the spreadsheet. The preview version looked better too, but formatting gets squished when restricted into the discussion template. As an alternative, I was thinking we could output the list of names and proposed rewards from step "H" (easier to parse) - which could lead people to investigate more closely if they see values appear too low/high - for themselves or others. |
Beta Was this translation helpful? Give feedback.
-
|
Well, an anonymous view of the spreadsheet is possible, so I don't think we'd need that at all. There would be that period on Saturday when people can add their remarks, I'm against any duplication in the process if it can be avoided. |
Beta Was this translation helpful? Give feedback.
-
|
Ah - my point was just that step H is needed for final list that would be submitted to the sender of rewards in any case. I was thinking it could be added to the discussion to start so people could see the damo values, and then they may be compelled to click to the spreadsheet if they wanted to see more and interact. |
Beta Was this translation helpful? Give feedback.
-
|
Oh, I see, yes, the list has to be published here anyway, so you can paste it immediately after proposing contribution levels. So, if you think that's ok, after the issues are collected you (the member who opens the discussion) can set initial levels approximately and then paste the temporary list here for community to discuss. |
Beta Was this translation helpful? Give feedback.
-
|
There had been three tasks in this cycle finalized by SCN9A (https://github.com/asastats/static-pages/issues?q=is%3Aissue+is%3Aclosed), we should add them too. Also, we should define a budget for opening this discussion and collecting issues and filling the spreadsheet with them, that's a one person budget and it could be the simplest to connect that with the creation of the final list (so we can obligate the person who opens the thread to create the final list and ping me). |
Beta Was this translation helpful? Give feedback.
-
I just added
A label is a cross-section property, I've just labeled all of my pull requests to archived, so please can you set the remaining three too if you addressed all the related rewards in this cycle. https://github.com/asastats/static-pages/pulls?q=is%3Apr+is%3Aclosed |
Beta Was this translation helpful? Give feedback.
-
|
Yes, I can do this - but to clarify - are you suggesting that we should track all issues AND all pull requests as rewardable events? Or is this just limited to pull requests on ~static-pages repo? |
Beta Was this translation helpful? Give feedback.
-
Well, I'm not suggesting anything at this moment, my only suggestion is that we discuss everything and that we should avoid prejudicing anything at any cost. So, I'm not suggesting that every/any pull request should be rewarded, but we have to allow a possibility to reward such a contribution. So I'll say again: this thread is a central point. If something should be rewarded out of regular issues that have been labeled as addressed (and that implies getting a reward) then that can happen for a cycle only if a discussion here brings such an outcome. And again, if the community misses something that happened in one cycle that absolutely doesn't prevent rewarding in some other cycle in the future. Therefore "addressed - archived" link instead of time-based filtering. So, I don't think your PR deserves a reward. A matter of fact is that I somehow rushed with labeling it But I am sure my pull requests there won't be rewarded so I labeled them archived in advance. Everything else should be discussed, but we simply - this is based on experience - have to define defaults somehow. And that's the reason why I've proposed that you (someone who opens a rewards thread here) spend some time contemplating about every contribution and define the initial level (B1, B2, ...). If no one complains, or better to say if nobody else participates in the discussion, or you don't change your mind, then that would be a source for the final list. And so we should have that predefined budget for a community member who opens a thread on Saturday. |
Beta Was this translation helpful? Give feedback.
-
|
Fair enough! My chief concern here is still trying to keep it simple. Investigating all pull requests takes it to a new level in terms of default rewards calculations - then as well figuring out if they are not yours + not included in listed issues...would be some investigation. But - to your point of saying that a pull-request that is identified by the community (maybe a nice text update w/ no related task submitted) - for sure if is reasonable to include it or note it via a discussion. |
Beta Was this translation helpful? Give feedback.
-
Well, I'm going to focus on this in the future and set the archived label for all of my pull requests. And for the others, if you agree we can present a new related label, a label for community members' work that we are not sure in advance it will be rewarded. For example that's not the case with those PRs that I left to you to label, they for sure deserve addressed label, I'm talking about some other examples so I don't prejudice with addressed.
Yep. And again, if you (someone who opens a thread) misses adding a PR and also a discussion doesn't reveal that, nothing wrong actually happens. Once when you or that contributor or whoever else points at that we simply label PR as |
Beta Was this translation helpful? Give feedback.
-
|
Since there have not been any additional comments or concerns noted here - I am wondering if it is time to output the associated rewards noted above from Ongoing Rewards per step H w/ the 2x coefficient? If so, I can also submit the steps used as a rewards-process.md doc draft - which can then be refined as we figure out improvements each cycle. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, absolutely, please proceed. I'll just add a remark that this week is an example, but in the next cycles the timeframe for the whole process should shrink from this week-long period to a couple of hours or a whole weekend at maximum. Btw, we're talking about a list file, but it comes down to pasting those values here in a comment, selecting that text and then clicking that Code <> icon from the toolbar. |
Beta Was this translation helpful? Give feedback.
-
Exactly, and I think next time - the person processing attaches this list in the discussion at the beginning, but then links to the spreadsheet. In this way, maybe it will catch a persons eye that some total reward is too low/high, and encourage them to go to link to review more. |
Beta Was this translation helpful? Give feedback.
-
|
Rewards list output: |
Beta Was this translation helpful? Give feedback.
-
|
Good question! I had originally meant it to be the rewards list output above from 8 days ago + the PM I sent you in Discord re: the review of pull request for the doc. This first try is a bit different than normal, as this discussion is both reviewing/discussing the process, as well as trying to follow it (in the same discussion). Assuming that the person who opened the discussion is the same person who measures the rewards and moderates the conversation (may not always be the same person) - then per the document in step H, that person should just contact you (or any future sender of rewards):
At the moment, Discord seems easiest - though maybe you prefer some other way? |
Beta Was this translation helpful? Give feedback.
-
Well, I'd probably always prefer Channel for everything ASA Stats related. In this case, that would be something like this: @ipaleka The list is finalized. |
Beta Was this translation helpful? Give feedback.
-
|
Works for me! @ipaleka the list is finalized! One additional note: to allow for latecomers to comment (assuming we get this process shortened to ~1-2 days) - I think it would be nice to wait to close the discussion until rewards are sent. Then, if some person has some issue after that point, they can approach with some new discussion. |
Beta Was this translation helpful? Give feedback.
-
The way how we had dealt with this before (not that there had been much of complaints tho...) is that we simply transfer everything to the next cycle. And that will work in any case except maybe for somebody's being rewarded too much for something, in all other cases there should be no problems |
Beta Was this translation helpful? Give feedback.
-
The rewards will be sent from tomorrow. |
Beta Was this translation helpful? Give feedback.
-
|
I'll use this opportunity to talk about Part 0 :), respectively about Ponzi's rewards system and a guideline for setting the levels for contributions. I'm bringing this comment as an example. Afaic it's an absolute maximum a community member can provide to the ASA Stats community. So, the initial commenter introduced the problem, and that comment by SCN9A brought absolutely everything a community member should comprehend in solving the problem, After that, crafting a solution is just a matter of effort and inspiration. I know how you AlgoiRhythm dealt with static pages work, defining the budget as a combination of multiple 100% and 50%, and I'm wondering how you would be approaching to this? So, the original commenter brought the issue and he surely should be compensated fairly for that. But, my position in this is that the reward for actual solution should be the smallest part in the whole budget, while the SCN9A comment should get at least 50%. Like:
What is your opinion on this AlgoRhythm? Do you share my position that the guideline, in this case, is absolutely desirable? The other thing that I want to emphasize here is the exact example where that guideline fits perfectly. But, from my experience in ASA Stats, if we advertise this too much among people who don't have SCN9A's enthusiasm and intelligence, then we can get an outcome of people copying and pasting bunch of unneeded and irrelevant materials and we end up in having much more trouble than in a case where we simply ignore such a guideline in a rewards process... |
Beta Was this translation helpful? Give feedback.
-
|
@ipaleka - to the point about the recent issue with FAQs and impossible total of 300% - I'm hopeful that we don't measure like that again in the future. That was meant to be a one time "hack" of a solution to properly reward all contributors to the FAQ - which covered a long span of time, w/ contributors, editors, etc. For all future cases: I am hopeful we can use natural math to = 100%, and adjust the coefficient and/or Ponzi's spreadsheet values if we feel rewards are too low (or high). As to how we (I guess mostly me) have measured participation in other cases so far? We have generally just done direct splits by number of participants in most cases, asking the recipients if some clarification is needed. To date, most have been quite humble and agreed to even splits. In terms of rewards from issues: based on the % listed here, I'd propose a split similar to the following:
Some questions:
|
Beta Was this translation helpful? Give feedback.
-
Ok, great, that suits me well.
Yep, we are pretty much a peaceful community... :)
In this particular case, he actually did as I moved that issue from Trello. And dunno about the real cases... My initial suggestion was that only the person who submits in GitHub gets rewarded, then people argued about that and we have discussed something like split rewards. Dunno really, the ideal outcome would be that the original proposer gets the whole Proposer's amount, but that comes down to me posting all the issues... We simply have to find a solution for this problem.
"Guideline" is a guideline to a solution: SCN9A brought the guideline for everyone on how they should approach the problem and how to solve it. And if we had got a punishment system then Sammy would have got it for his comment(s). I pointed out the first time to SCN9A comment after realizing he hasn't read it, and then he just skimmed it and provided another lame comment. We haven't got those things for a while, but those two are examples of what we politically correctly call "paved with good intention".
In this case, the solution is that crafted text "ASA Stats evaluates and presents ASA and NFT price information on the Algorand blockchain. Protfolio Tracker, API and Feature-Driven Development. Check it out.". Deployment is just a side-effect of getting that solution. In the pre-DAO phase, all the deployment is done by the Team so we don't need additional issues. In the DAO phase there should be one or two other issues that would be created after addressing this issue, having the task of merging the text with the related page template and the task of provisioning/deployment on the website. |
Beta Was this translation helpful? Give feedback.
-
|
Btw, I'm about to close issue #67 without setting any additional label - the issue is brought by me and so there are no rewards involved in the process. |
Beta Was this translation helpful? Give feedback.
-
|
We still have to discuss rewards for the moderator(s). What is your opinion @AlgoRhythMatic, should we go with flat rewards for a cycle rather than calculating them from the allocated rewards in the cycle? |
Beta Was this translation helpful? Give feedback.
-
Well, that would be against our voting. We have the option to set the rewards for the moderator to any amount we want without making people disappointed. My only concern is that nobody would be eager to take that step on Saturday. And tbh, I'd rather see a fight for 50 times bigger rewards than confirming you're the only one who does it. |
Beta Was this translation helpful? Give feedback.
-
Well, for me it always comes down to time spent...how much does each person consider their time worth? Based on Ponzi's model, we tried to break this down into levels per amount of hours spent. My current estimate is maybe ~3 hrs for a person who has never performed this operation, and maybe 1-2 for a person who has. Not sure how to quantify the moderation parts...coming and going + interacting with questions, etc. I'd be interested in your measurement, or others feedback that would be interested in trying it out (@rom1henry, @Babbexx22, @SCN9A, @bear1bear2bear3 or anyone else from the community). I expect that whomever tries it next may need a little help, which could improve the process, etc. |
Beta Was this translation helpful? Give feedback.
-
|
Just to be clear, and I'm writing this for the others, I don't mind you always doing this (afaic you're the perfect choice), the meaning of "only one" from above is not because we're going to get bored of you doing this, I'm rather afraid that you become bored of doing this. And also that this is sort of dull work and we need you where the fireworks are. :) |
Beta Was this translation helpful? Give feedback.
-
It'll be good that someone else tries his hands on it...with your supervision.. So we can have a range of able hands, should you be busy at one point or the other. |
Beta Was this translation helpful? Give feedback.
-
Exactly! It is expected that AlgoRhythm would be able to perfectly craft the rewards document after a few sessions with other community members. |
Beta Was this translation helpful? Give feedback.
-
|
The following initial DM will be sent to those community members who haven't been rewarded yet: Hi! I'm an admin in ASA Stats and I want to ask have you read our rewards thread for the previous cycle, especially this comment #44 (comment) ? After response, they will be asked to opt-in for ASASTATS and provide their public Algorand address. |
Beta Was this translation helpful? Give feedback.
-
Overview
The purpose of this discussion is twofold:
Part 1
Rough draft of the process used get to part 2. After discussion, the final process definition will be stored in a Gihub rewards.md:
a) From a web browser, navigate to the "Ongoing Contributions" tab of the ASA Stats Contribution Rewards spreadsheet. Enter a new "Period below" header by cutting/pasting an entry from the previous cycle, then adjust the date range and name of community member tabulating rewards.
b) From a 2nd tab, navigate to the ASA Stats Github Channel Issues page:
https://github.com/asastats/channel/issues
c) From Github tab: use a filter to search for all closed issues w/ the "addressed" tag.
Example syntax for a cycle with end date of April 29th:
is:issue is:closed closed:<2022-04-30 label:addressedLink to this query: https://github.com/asastats/channel/issues?q=is%3Aissue+is%3Aclosed+closed%3A%3C2022-04-30+label%3Aaddressed
d) From Github tab: click on each issue to review details. Use the linkage in the description to determine the related recipient/s. Use the Contributions Types and Levels chart to determine the reward type and impact value to add to the spreadsheet.
e) From spreadsheet tab: enter in one line of data per reward recipient. Fill in the appropriate field values in columns A through I for: Person, Period begins/ends, Platform, Channel, Contribution (reddit, discord, twitter or Github link), Contribution Type, Impact level, and % of reward to receive.
f) From Github: as each issue is added into the spreadsheet - remove the "addressed" label, and add the “archived” label.
g) Open a new Github discussion w/ a descriptive title like: Discuss and review the rewards cycle from 3/26 - 4/29. Add in the chart of the current rewards for community review, allowing for feedback regarding impact levels. Make adjustments to the spreadsheet as needed based on discussions.
h) From the spreadsheet: after the community has had sufficient time to comment and finalize all reward impact levels, navigate to the "Rewards by Period" tab, and copy the final output values from column D into a list in a text file. If there is some adjustment coefficient (currently 2x in this cycle), multiply values to adjust. This output list should then be added as a comment in the discussion, and also provided to the official sender of the rewards.
Part 2
The following rewards entries were created using the draft process from part 1. Per step "G" the entries are presented here for community review and discussion, but may also be reviewed from the bottom section of the Ongoing Rewards tab from the spreadsheet. Please review the entries, and comment to discuss if you feel any of the Impact Levels should be adjusted per guidelines in the Contributions Types and Levels chart.
Beta Was this translation helpful? Give feedback.
All reactions