Some initial thoughts on the App Mining Program

Also I very much agree that Misthos is amazing from a fundamental perspective. It is really fascinating and could have seen it be higher ranked. It may have been that the ability to fully understand its fundamental nature is difficult for the common person. It is definitely something to think about when thinking about what things to incentivize: Apps that people love to use.


You and others hear me well Patrick. You guys knew that I think app mining is fail, currently. And I want a better solution for that.

I strongly disagree with this statement --> 1073 votes for Misthos vs 31 votes for Coins on Democracy Earth. Common person have better judgment than “PH expert” in this case.

As I told you yesterday, I think users should have more than 50% voting power. And I think delegated system (like EOS) is bad. And Blockstack should not follow that path.

I also told you yesterday, it’s not only my questions. It’s from others as well. And I’ve to answer questions like,

Why voters don’t need an Blockstack ID or using app?
Why Product Hunt in charge?
Did Blockstack PBC controls everything?
Can Developer live with app reward?
What if the community don’t like a Blockstack’s decision? Can we stop it?
And many more.

I don’t wanna talking private to you because If many people know the problems then we could have many people think about it. Not just you and me.

I only support Bitcoin and Blockstack. My reputation tied with Blockstack. I just want you and others can find solutions.


First I think we should set up a consistent way of getting feedback from the community, lest only one or just a few people’s opinions be heard. That’s not to say your opinions are unexamined ones, as there are some good points I’ll address in the paragraphs below. Currently we have collected a survey from app creators themselves and will publish it once the results are in. We could do something like this for the broader community, however that is defined, where thoughtful questions are posed and we tackle them methodically to find solutions. What you bulleted above was the most helpful thing so far. Let’s mark this as an action item to develop a more consistent and methodical way to gather community feedback.

Secondly, now is the time for experimentation as this is a currently just a pilot paid out in bitcoin and not Stacks. It is safe to assume nothing is permanent but also that this is the time to let things play out and see improvement in App Reviewers by giving them feedback rather than making hasty or reactionary decisions. We still need to learn and that takes time.

On the community having more than 50% voting power, we should consider why that or even 100% of the vote would be a good idea or at the very least not get gamed. In this experiement, also as outlined in the token white paper that stakeholders have accepted, is that the expectation is to have many app reviewers will exist to avoid gaming of the system. The number is yet to be determined, and I’m betting we won’t have an answer right away. App reveiwers serving different functions could presumably create the conditions for a less gameable system. That’s what we’re experimenting with here. There is a reason to not simply have the community do all of the voting, at least in the early stage of things. We are in a stage right now where we get to test things out, so let’s do that methodically and patiently so we can learn over a series of months. There are no limitations to what types of specialized entities could become app reviewers or how many of them the system needs to achieve something that is better than say a 100% direct community vote. Let’s seek out the best options and improve the current ones.

Let’s go through some of your questions:

“Why voters not need a Blockstack ID or use app?”
Presumably people are using apps, but likely not in every case. Ideally people vote on stuff they are well informed about, but just like in real world political voting it’s not always the case that they are. This is an incredibly tough problem to solve but I’d say at first approximation the general field of apps were ranked fairly well with some debatable ones. Personally I was thrilled it wasn’t completely off at the start of this alpha experiment…that was a great sign. The only way you could enforce folks to apps every time they vote would be to conduct a like experience where people provable review the app by sending in a video. Or tracking the user somehow, but it’s not clear that that is a good solution. For Democracy Earth they are planning to add Blockstack Auth to it, which is exciting.

“Why is product Hunt in charge?”
Product Hunt is currently not in charge. They are representing the most influence they will likely ever have at 50% vote weighting. As more app reviewers are added, one could imagine any one App reviewer only holding something like 1-20% weight in app reviewer votes, this would include product hunt if they remain an app reviewer. Their final rankings are represented by their community and internal team. To quote you “the More Users —the more people wanna support the system”. Between product hunt’s success as the pre-eminent platform for reviewing products and their reach to users of products currently attracting 7,000,000 monthly visitors, they seemed like a great fit. They are also quick learners, and I expect them to improve as the program goes along. After the pilot ends, if you don’t like them, you as a member of the community can vote them or any other app reviewer out via election. During the pilot program Blockstack PBC can act on behalf of the community for speed and distribution purposes. This will not always be the case as Blockstack becomes less as less involved through the community vote.

“Did Blockstack PBC control everything?”
Not at all, for example right now the community vote represents half of the weighting for ranking apps (not including if you vote on product hunt). As far as being a steward of developing App Mining during the pilot stage, it is indeed taking on that role. It’s a role it wants to get rid of as soon as the pilot stage is complete. There is a trade off of coordination versus decentralization and the goal is to arrive at thoughtful decentralization as soon as possible.

“Can developer live on app rewards?”
Yes, during the pilot phase of you happen to be in the he top 5-8 placers you could pretty easily live off of that. In a world where the monthly rewards increase even further, you could fund full time development of many more app teams.

“What if a community member doesn’t like a Blockstack decision. Can we stop it?”
Absolutely. The idea here is not to commit any self inflicted wounds with app mining, and to proceed very carefully. That’s why we’re are ramping up sending surveys and sharing the results with the community, and having discussions out in public. The reason to designate a pilot phase for app mining is to also develop a more formal way for setting up the real thing for the long term. This will require more voting, checks and balances, proposals, and thoughtfully moving away from central point of control.

Lastly, regarding me asking to speak to you on a call, I don’t want to talk privately about anything related to app mining, but rather talk to you personally about how you’ll best be heard in the community because I do care about you being heard. I think as the community grows, you’ll need to seek a better approach to surfacing issues. I’m getting a lot of feedback that your tone is not being well received by community members on the forum and on twitter and just want to flag that for you, whether you believe it or not. I think what is most beneficial for your involvement in app mining discussions it to continue to list items like you did with the bullets and dispassionately propose better solutions and a well-examined reason why it’s good (i.e. how it is less likely to get gamed, etc). Better decorum in these forums and on twitter is constructive and can help decision making because it avoids people tuning you out because of the way the message is delivered. I understand you are passionate and care, we all do, but we must be respectful to each other by seeking to understand first before seeking to be understood. Those are my rules for myself and I’ll continue to keep an open dialogue with you respectfully and thoughtfully as long as you promise to do the same.

Thank you again for your questions Loc. Shortly I’m planning to post another forum thread about what new app reviewers we should consider. I hope to see you in that discussion and in the survey results with some great ideas.


Patrick, you write a lot. I don’t want go through all your stuffs because it’ll be too long long long. But here is some in my head, I’m very disappointed to some of you who avoids admit being wrong.

The coins dev is lying right to my face that he didn’t know about the bug, even though we’re talked before and he asked me take a screen shot as well.

And whoever makes PH voting weight more than 50% know that it’s true - PH control app mining now. We have two parties:

  1. Democracy Earth Votes Misthos win.
  2. Product Hunt votes Coins win.

In the end, Coins win. It’s very clear to me that PH voting weight more than 50% and PH control app mining for now.

TL;DR: I’ve other audiences as well. If (s)he can’t taste my expressions, I’m ok with that.

I support Blockstack about a year or so; more than 64 videos about Blockstack on Youtube, a Vietnamese language website which spreading Blockstack, organizing about 7 Blockstack meetups (one meetup in the no.2 university in Vietnam - students ask question about Blockstack vs EOS) and I’ll attend a coming meetup on Nov 10 as Blockstack Community member and Moderator for the event. That said I’ve skin in the game.

Since a few months ago, my audiences asked me to ask Blockstack that “Does Blockstack voting mechanism same as EOS?”. I sent this question to Ryan, Jude, and Larry privately. Larry told me that he’s not working on that and he couldn’t give me an accurate answer. So I posted it on Slack later. No one answer me…

Anyhow, I promised to my audiences that if I see “Blockstack voting same as EOS”, I will make sure anybody know that I against that idea and If Blockstack still choose that path. I stop support Blockstack.

By make sure, I meant my audiences can verify it. Some of my audiences use Twitter, Facebook, Youtube. Long story short, I don’t care whatever other think about me. I do my best and that’s enough.

1 Like

Thank you so far to Thomas (@digitalwaveride), Loc (@nguyenloc), Justin Hunter (@jehunter5811), Prabhaav (@prabhaav), Justin Carter (@bodymindarts), and Dan (@dant). This thread is extremely valuable to everyone in the community and to the Blockstack team working on making App Mining a success.

I want to summarize your major concerns voiced in this thread to make sure I have a correct/complete understanding. Please correct or point out anything I may have missed:

  • Add an App Reviewer that ranks apps based on objective, performance-based metrics.
  • Increase transparency on how App Reviewers rank so apps can improve in the rankings.
  • Require all App Reviewers to have a Blockstack ID and try the app before ranking.
  • Add the requirement for Apps to use Gaia in order to be eligible for App Mining.
  • Increase the amount of influence Democracy Earth rankings have on final rankings (currently 50%)
  • Apps eligible for App Mining must be actively maintained.
  • Give app developers ability to have some say in the rankings via a vote.
  • App developers receive feedback in addition to a ranking each month on how they can improve their apps and potentially get ranked higher next month.
  • Add different categories for apps in App Mining. Right now the program rewards consumer apps specifically but may be more difficult for types like B2B.
  • Make the distribution of rewards more concentrated so the top apps get more money and can actually sustain development off App Mining.

My goal here is to simply collect all the input and make sure nothing slips through the cracks.

The time you’re taking to help improve App Mining is really appreciated and please keep the feedback coming. There is no better way for us to improve than to have open, honest discussions like these. Thanks :pray:

– Muneeb
CEO, Blockstack PBC


Wow - lots of great discussion in this thread!

I’m going to add my take on what App Mining does, reply to some earlier comments and add my 2 microstacks.

What App Mining does

Without users in an ecosystem developers won’t build apps. Without quality and useful apps, users won’t join an ecosystem.

App Mining attempts to solve this chicken & egg problem by providing incentives & resources for people to develop new, high quality, useful apps that will entice users to enter and stay in the ecosystem and a distribution mechanism to get the apps noticed by potential new users.

Resources reduce the financial risk of building in new ecosystem and extend runways giving teams more time find product-market fit, funding and a sustainable business model for their app.

Incentives push apps to have the characteristics and values the community wants.

Distribution helps apps and the ecosystem get more users.

+1 Apps that don’t work should get disqualified and no money.

I agree with this - this is the approach is what I plan take with the new standalone browser entity.

App Mining is NOT a long-term business model - nor should it be. Ultimately the people best suited to determine if an app is valuable are the people using it. They will do this by paying for the app in some way.

I agree that this process needs to be simple and 100% transparent.

Yes. Ultimately users of Blockstack apps are the most important constituency.

I see Product Hunt and Hacker News - as distribution components of the app mining process. They get apps out in front of a wider audience. Agree that the actual votes of people shouldn’t weigh too much on the outcome.

Some apps are well-suited to popularity contests on Hacker News or Product Hunt. Others aren’t. Other apps - industry-specific enterprise apps for example - might be highly ranked by an app reviewer that is an enterprise-focused VC fund.

We would do well to consider finding a more diverse group of viewers and switching from a single aggregate ranking based on all reviewers to a ranking for each reviewer.

You get what you pay for.

To be eligible for app rewards mining your app should need to:

  • work
  • have the characteristics* that meet the values** of our community.

I sense that some of the frustration in this thread comes from the fact that the qualifications and voting system aren’t in line with those values. And we’re putting a great weight on the views of people that can’t even be bothered to join our community.

We Blockstackers are a bunch of people that are unhappy with the current internet and want to create a new one aligned with our values, yet it appears as if we’ve handed control of a large part of resources and the future of our ecosystem to a bunch of people (users of Product Hunt) who can’t even be bothered to try out the apps.

It’s like giving North Koreans control over how Canada spends its budget. If North Koreans want to have a say over how Canada works, they should at least go move there and join the society.

Blockstack token holders are invested in Blockstack - they have skin in the game - they care, so of course they tried out the apps. If you try out the apps, you know that Misthos works while Coins doesn’t.

Blockstackers that understand our vision and share our values also understand that Misthos is aligned with our values and see its potential. Most Product Hunt users don’t share our values - they’re probably the same people that laughed at Bitcoin years ago because they didn’t get it.

I think it’s valuable to get Blockstack apps out in front of a wider audience via channel like Product Hunt because ultimately we want to sell them on our values via benefits that they’ll enjoy from apps in our ecosystem. I think it’s also okay to give Product Hunt performance some small influence on the rewards paid out. However, the vast majority of the weight should be given to app reviewers that share our values and have skin in the game.

** Here’s my take on our ecosystem values:

These values imply apps with certain characteristics:


And on the subject of the Coins Dapp, haven’t the creators of this taken advantage of the Blockstack bounty system? Correct me if I’m wrong (and if I am I 100% retract what I’m saying here) but they created a product for the bounty, collected first prize and then ceased to do any future work on the Dapp. I thought the point of bounty rewards system was to kickstart some of these Dapps, but if Coins is a predictor of future performance with the bounty program it appears the devs did the bare minimum they needed to create the Dapp and then vanished from the scene once they collected their payout. I haven’t seen one update since it was featured on the homepage. However, I do remember seeing some mockups for future improvements in Coins, but to my knowledge this hasn’t happened nor I have seen any proposed updates.

I bring all this up as a precursor to this question: Is the Blockstack Bounty program incentivizing the right type of devs? Is it only going to bring people into the space to collect the bounty and then vanish after the rewards have been announced? I haven’t combed through each one of the portfolio Dapps, but I hadn’t seen any updates the last time I checked. Neither CoinKitty or Coinfort work, and Coinstack hasn’t changed anything with their UI.

Sorry, I realize this doesn’t have much to do with the App Mining program, but my concern is that people might try and rush their Dapp or create a Dapp simply so it can start accumulating mining rewards without there being any real plan to maintain and improve the app. I realize this is harder because the rewards are put out monthly and the top Dapps do indeed have active devs behind them, but I worry that the larger point of incentivizing people this way might not work in the near or long term.

But my overarching point has to do with the Blockstack Bounty program and in turn the fact that there are Dapps listed that don’t have much value in the Blockstack ecosystem.

As a fix action, instead of paying out the bounty all up front (which is how I believe it’s done), there should be a stipulation about updating the Dapp and maintaining it in order to get the full payout. So maybe they get paid $10,000 up front and then $2,000 a month as long as their Dapp is maintained and updated.


I agree with this sentiment and think I’ve made my thoughts on this pretty clear with my proposal in this post: Proposal for App Mining

I think the bounty program was largely a failure. Stealthy came out of it, and Stealthy is doing great. Techcrunch Disrupt, tons of press, continued development, etc. However, other apps to win the bounty did seem to take the money and run. Coins might be another exception. I know Thomas is still working on the app, but I don’t know how much has been deployed since it won. It’s probably important from a transparency standpoint to note that both Coins developers worked for or are currently working for Blockstack. I’m sure time is much more consolidated to Blockstack items than Coins items.

But Kit’s point is well taken when you look at Dappy Wallet and whoever won the Slack clone bounty. I’m not sure if the Slack clone app actually got paid, but if they did, they haven’t even released an app that anyone knows about.

The goal with App Rewards should be to avoid the mistakes of the bounty program (which I believe is dead probably for the reasons Kit outlined). This thread is raising all of the right points to help make sure App Rewards works better than bounties ever did.


Hey Everyone! ! :wave:

Just for those of you who don’t know, I’m the maker of Coins. Figured I should step in and say a few things.

First off, I’m super happy that there has been so much great feedback in this thread. I think App Mining is super cool, and I’m happy that it’s happening. When I submitted Coins to the program, I had zero expectation that it was going to place even remotely where it did. I hadn’t really worked on it much in the last months or so, at least not the version that is live.

I understand that there is a lot of interest around Coins and the place it got, I’m going to try and talk about a few of the points that stick out to me:

And on the subject of the Coins Dapp, haven’t the creators of this taken advantage of the Blockstack bounty system? Correct me if I’m wrong (and if I am I 100% retract what I’m saying here) but they created a product for the bounty, collected first prize and then ceased to do any future work on the Dapp.

This is not really correct. I built Coins with a good friend of mine. We split the bounty and started work on a new and improved version. Some stuff happened and my friend took a job somewhere else and left me in charge of Coins entirely.

Since winning the bounty, I’ve worked on two new versions, one that was built with React Native, and another one I started after I realized it wasn’t a good idea to build the react native version. I’ve easily spent hundreds of hours working on various versions, learning more about how to do things correctly, and just having fun writing code. I presented an early prototype of Coins v2 at Blockstack Berlin, and after that I tried to seek funding options. The app started getting expensive to work on because I wanted to use an API that costs hundreds of dollars a month to subscribe to. I could not find funding and so my work on the new version of Coins essentially stopped (I was working full time on other things at the time).

App Mining came along and I was like, cool, maybe this could fund the API that I wanted to use, so I submitted my app. It has now placed and I’m going to use the BTC it earned to pay for a year-ish of the api!

There has also been some concern over the bugs in the current version of Coins. I’m going to be completely honest, there have always been some bugs in the app. It was built quickly and when I first started working on V1, I had not really written or touched react before. There were always bugs that certain people had that I could not replicate. @nguyenloc It seems like you feel like I’ve wronged you or lied to you. I apologize that you feel that way, it was never my intention. I mostly avoided these conversations because I didn’t think anything I said would change your mind :slight_smile: When you posted about the bugs, I had looked at them and could not reproduce them at the time. I was also hard at work with a new version that I thought I would ship soon, so I didn’t think it would be worth it to spend time trying to debug a problem I couldn’t recreate.

I’m actively working on the new version, and hope very dearly to release it soon because I think it’s going to be helpful for people who want to track any kind of digital asset. It’s a progressive web app, too, so it’ll work great on mobile devices :slight_smile: Here are some screenshots from the new version:


What a transparent and thoughtful answer. I do still standby my earlier concerns about the bounty program and how that could also lead into the app mining rewards program (the other token Dapps still haven’t been updated or just flat out don’t work) but you provided lots of clarify about Coins and the future of it with your answer.

You are clearly a talented dev and it’s exciting that you’ll be able to update Coins with the addition of the app mining rewards to your financials. All the best!


Thanks for the clarity and acknowledgment It was not my intention to steer this discussion into the usability of Coins, but rather to highlight the problems with the App Rewards program and more importantly the mysterious PH voting system.

I hope this and future discussions will help Blockstack and the eco-system grow in a positive manner.


Sharing some thoughts on behalf of Richard, Felix and the team at Afari.

(1) Congrats to Blockstack on a smooth App Mining Pilot!
First, congratulations to Blockstack PBC for actually pulling off a smooth first run of App Mining. It’s a massive effort and Afari is excited to be part of future iterations of the program and hit that #1 spot soon :wink:

Moreover, thanks to everyone for sharing their concerns and points of view above. It’s great to see such interest and emotions shown in how app mining works, as that shows clear demand for it!

(2) Concern about Democracy Earth / Stacks Token voting + Associated Scores
Our only concern from the alpha run was with the Democracy Earth voting portion of the voting. First, it’s important to note that only 17 voters actually voted in the Democracy Earth (DE) portion of the vote. We can all agree that we’d like to see this number increase, so we’re curious to know what measures are being put in place to encourage STACKS token holders to vote in app mining?

Secondly, we’re curious as to why both a likeability (total upvotes) and a traction (net upvotes) score are being used when they’re fairly similar metrics? It would make more sense to have an either total upvote score (likeability) or net upvote score (traction), so why have both? What sort of game-theoretic scenario did you have you in mind that you’re trying to prevent by using both scores?

Moreover, what’s the purpose of having a downvote option? Is it for people to downvote apps they hate? Surely the better option is to simply not vote for that app and allocate your votes elsewhere?

(3) A Call for Transparency on scores for each category
Lastly, we wanted to make a call to make all intermediate scores (i.e the PH community, PH team, DE likeability and DE traction) and their z-scores available for for the app mining participants to see which things contributed to their final score the most. We understand that 3 out of those 4 can be computed already but it would be nice to have all 4 so that we have complete picture.

Onward and upward,
Team Afari


Like I said before, I raised questions because I want you guys (the whole community) to find solutions. Since everybody know there’s some errors, we can work things out.

I don’t wanna comment on person A, B or opinion A’, B’ anymore because maybe I just see a tip of the iceberg and you guys have good views than me - can see clearly, wider and deeper.

The main point, we know what we have to do. Go Blockstack!


Hi there!
I’m a founder of BC Net who created Dappy Wallet. Looks, it is my turn to add few things.

First, we decided to make Dappy Wallet thanks to BlockStack bounty, and, it is a product which we have been working on since that. Bounty was a trigger for us to start a new project together, and till the bounty results, we had absolutely equal chances with others to win. As a result: we made a product which needed for us as users and for BlockStack as a community.

Last few month we had a chance to review what has been done and add new features like superlative security in storing private keys and transaction signature, add support of new currencies, UX improvements and others. We have a tough plan to follow and very interested in feedback from users and developers.
At this point App Mining gave us even more: we’ve got more users who sent us bug reports and change requests, plus we’ve get paid and me with team consider it as flag that it is not a simply “yet one more” wallet, but app which is interested by relevant people: community, app reviewers, BlockStack.

My point is both AppMining and Bounty “were” and “are” vital for us in creating a new product.

So Dappy plan is to release new enhanced security feature, cash-in/out partner integration and many UX updates.



One other question for @patrick and @muneeb
Is there a clear criteria in terms of target platform these apps should be developed for? I personally try very hard to support desktop and mobile. Some apps are desktop only. Some are mobile only. Is there any clear understanding of the value of the different platforms with respect to App Mining rewards?

If the reviewers only review a desktop version then I should focus my work there, right?


Hey, it looks like we haven’t properly communicated what these two scores are.

“Likability” is actually “net upvotes”. The score is ((upvotes - downvotes) / total_votes).

“Traction” is just “total votes” (including downvotes)! Meaning you can improve this score by getting downvotes (although it’ll hurt your likability score).

This was created by the game theorists, and I think it’s actually brilliant. If you look at a real example, take Afari and Coins.

Afari got 803 upvotes and 50 downvotes. “Likability” is 88%.

Coins got 31 upvotes and 0 downvotes. “Likability” is 100%.

So, Coins does better on the likability score, but just slightly.

On the traction score, Afari got a much, much better score. I think most people would agree that they should be rewarded for many more votes. Voters have shown more interest.

So, why include downvotes in the traction score? I think this footnote from the game theorist’s paper (which we will publish when finished) explains it best:

Perhaps a concrete example from the online video gaming world may be illustrative here. Bungie’s famous 2014 game DESTINY has generally received subpar reviews from the critics. (The Metacritic score of this game is, for instance, 75/100.) And yet in 2014 it was the third best-selling game in the US. (By the end of 2015, DESTINY was purchased by 20 million users worldwide.) By contrast, FromSoftware’s 2011 game DARK SOULS has received raving reviews and obtained several “Game of the Year” awards. (The Metacritic score of this game is 90/100. Indeed, DARK SOULS is presently ranked by PC Gamer as the fifth in the "Best RPGs of all time” list.) And yet, this game is famously difficult, and as a consequence, with regard to mere popularity, ranked way below DESTINY. (It sold about 2.37 million units worldwide.) Deciding which of these two games is “more successful” obviously depends on the criteria with which one chooses to assess “success.”

1 Like

Thanks for the clarification Hank. We totally misinterpreted what the traction and like-ability scores meant. Perhaps it might be good to include that example in the Algorithm description post to save future misunderstanding. :slight_smile:

As you noted, the two metrics measure different things are are equally valuable as indicators of a Dapp’s success in this ranking algo.


Whatever platform works best for your app tbh. Afari is moving toward a mobile first approach.

1 Like

App Mining Improvement Proposals are moved forward in the App Mining GitHub Repo: .