Hey guys, super excited about the launch of the App Mining programme. Congrats to the launch! Really appreciate it
Reading about it this morning here are a few spontaneous, initial thoughts… please don’t take them as a critique, but as a well meant feedback. This is a real hard problem to solve, you have my full empathy … The following are just my 2cents from an app developer who is soonish releasing a first app version:
1) Upfront I was wondering about what are your goals, intentions, behaviour that you want to incentivise with this program? What are your OKRs?
- New signups?
- New app launches?
- Daily active users?
- Positive impact on the developer community ecosystem?
- Driving outside awareness of Blockstack?
- Incentivise the production / ux quality of apps on the platform?
- Incentivise longterm app projects?
- Offering a sustainable revenue model?
- Offering a model for Bootstrapping support?
Unfortunately I haven’t got the chance to read any public discussions about these intentions, yet.
Where is your focus?
2) Distribution of payouts:
"The highest-ranked app receives 20% of the total, the next highest-ranked app receives 20% of the remaining 80%, and so on … "
When I read this, the main signal of this distribution model to me was: Internal competition. Especially cause it is based on “ranking” not on total performance numbers.
Does it make sense to incentivise competition inside of a community? In such an early stage? What is the thinking behind this?
Would it make sense to focus on channeling all energy into competing with the centralised world? Incentivising the developer community to organise and help building each other?
So far I really enjoyed the community here at Blockstack … and I couldn’t have done many things on Textlibrary without the help of so many of you … @jehunter5811 @jude , looking at you… and all the others…, thanks so much! Love you guys.
But there is another equally important point that speaks against internal competition on the promotion side: If you look on classic dynamics on fast growing communities like for example the YouTube Vlogger world or also Podcasters … The most successful promotion tools they have is introducing audiences to each other. Having guest appearances on each others channels, frequently mentioning each other … doing collaborations etc. Promoting the “field” itself.
Imagine the revenue of these guys would be linked to a zero sum game ranking, where they compete head to head?
My question therefore would be, what are the reasons to not link things closer to a performance rating?
Trying to align the incentives of the platform with the reward?
Isn’t that the beauty of token marketplaces?
Maybe to early to launch this, though then why not considering a more “referral based” model, like for example numbers of BlockstackID signups that come in via an app?
Any other traffic metrics?
Sure, there is the danger of people gaming it… though BlockstackID signups (hard to fake, as costs involved)
If you would find key metrics where incentives are aligned, even if players “over-optimize” they at least help going into the right direction.
- What’s the potential life time value of a new Blockstack User?
- What is potential “revenue” of an engaged user (if experimenting with a retention metric)?
If you are afraid of a “peak”/“maxing out” problem you might consider to set a maximum revenue number per month and adjust obviously in the next round and learn from the issues be it black hat behaviour or too much growth.
The over all goal of the incentive program should be to create “a flying wheel” … not sure if a zero sum game ranking system is the best option here.
-> What can we learn from crowdfunding platforms like Kickstarter?
-> Rev Sharing models like YouTube?
-> Gig Economy marketplaces like Uber?
-> Referral programs like Amazon?
-> Patron programs like Patreon?
One additional comment on the ranking:
Depending on what your are intentions are (discussed above), if supporting bootstrapped projects is a goal, think about different dimensions (categories) for the ranking. After a while the most established companies on the very top of the list might not “need” the “extra bonus” for the top positions. As network effects are in place, the biggest platforms will stay on top.
Therefore maybe it might make sense to have a “rookie” category, where new projects can drive till they max-out a certain amount, or a certain time period.
3) The "Ranking Review"
This is a personal, spontaneous first reaction, a gut feeling, sorry for being honest: The App Reviewer Model feels very weird. I didn’t get it.
You mentioned you want to make the process of this ranking transparent, though you also mentioned that you don’t want to reveal all of it, so that it can’t be gamed.
You did a first alpha pilot round, though you didn’t share the reviewing process in the announcement. Only later I saw the blog post on the PH blog… in that particular post with pretty broad criteria to be honest (felt like American Idol?) … “Why giving this “power” to third parties?” … was my initial thought? - Then after digging deeper in the other posts and pages I found more detailed explanations how these reviews suppose to work… though to be honest things are not really easy to understand. And even if these reviews are based on metrics based not on a editorial teams, the elephant in the room is: What has success on PH or being liked in a DE community to do with the positive impact on the Blockstack ecosystem? I assume there is a correlation… though is it as strong as for example generating new Blockstack ID signups and therefore traffic in the Blockstack app store etc… and more importantly, why should someone at PH define these criteria? This is all really hard to grasp.
Maybe one of the reasons for my gut feeling weird.
Grants, Incubator programs, the VC industry… there are already a bunch of models in place in order to support projects (“rank projects”) based on metrics and analysis methods that are defined by an entity behind closed doors, based on subjective metrics and (hopefully) a lot of expert knowledge.
Nothing wrong with these. They have a track record.
Though I am wondering why a decentralized platform is choosing to go that route … even if it is just for the first steps… of creating a new model, that essentially tries to be a lookalike decentralized incentive model.
The “lookalike” starts with the naming … “App Mining” … in that case, not sure if it is appropriate. A “proof of work” that is not completely transparent and in the hand of a trusted third party … as mentioned, feels weird.
I am assuming there are a bunch of reasons why an engagement token marketplace is not yet possible, but personally, I would appreciate in such a case much more an active “YC program for dApps” … your Signature Fund program looks very promising…
To wrap this up:
Again, please don’t get me wrong. This is a very hard problem. Just trying to give input with my subjective opinion as an excited Blockstack developer
We all agree, that we don’t want to have “beer apps” and we don’t want to give subventions to click farms… though we also probably don’t want a subjective contest either, that is “camouflaged” as a metrics based ranking. Though most importantly we all want a system where all our incentives are aligned. It’s so early in this stage of “finally fixing the internet”, let’s make sure we all have only reasons to do this together.
From a quality standpoint in most cases charts and rankings seem to be flawed (just have a look into the app charts of a platform of your choice), and all charts are manipulated if they are powerful signalling tools (look at the NYTimes Bestseller lists). It will probably stay a never ending optimisation challenge around fairness, transparency and understanding of the intentions and incentives of all parties involved. No one should expect that we find a perfect solution in the first iteration. I guess what I try to say is:
Dear Blockstack team, you are doing a great job! Thanks so much for supporting the developer community.