Comments on February App Mining Results


As I can’t take part in the conf call later today, here are my comments on the results. Please do not take them personally, I am trying to understand why OI Calendar scored second last and improve the algorithms. I am happy to expand on them personally if you like.

  1. It is great to have a public github repo for app-mining!
  2. The algorithms failed to detect IMHO that is not an “eligible” app. It is a great project but with regards to blockstack it is just a hello world showing the user’s identityAddress.
  3. Algorithms failed to detect that Gladys requires to buy/have a Raspberry pi. How did TMUI test that? Is this “eligible”, open to all? Did I miss anything?
  4. Judging by DE is not transparent, from the outside the results look very random. Interaction between judges and app publishers on the DE platform did not really take place.
  5. From my perspective and looking at the top 20 apps, ~40k$ was given to 9 projects solving 4 problems. See here for details. While competition is good this currently prevents me from using e.g. a blogging platform because I don’t know which one to trust (too much effort to evaluate all 4?), which one is the best, which one will evolve and the next is already coming…
    Taking it a step further, how would 5 clones of OI Calendar score in the app mining program?
  6. The algorithms do not favor apps that deal with personal data (criteria: part of e.g. Google Takeout). E.g OI Calendar, MyWhereAbouts,
  7. Enhancing open source apps with Blockstack and publishing them breaks on Product Hunt. E.g. and are the same apps one without and one with blockstack. They have different upvotes. (Seeing the standardization procedures and priority list of riot/matrix, it might take 12 months before they merge my pull request)
  8. Ourtopia did not improve their app and went from 27 to 38. That is good (no improvements should be penalized) and that is bad (a decentralized community site is a wanted app IMHO).

From this subjective observations I started to file issues on the app-mining git, feel free to comment there.

Blockstack February App Mining Progress Review