TLDR: On the scale of 1 to 10, with how much certainity do I present the articles on this website. Highest confidence gets the value 1, and the lowest confidence gets the value 1. In some rare cases,Null confidence gets value of 0.

Fighting bias with yourself

I learn alot of things simultaneously, and being a jack of all trades is dangerous when you deliever it to someone who is an ace.

Confidence metric is something that I stole from Gwern 1 .

Whatever article I present here, I present to satisfy three purposes

  1. Developing a knowledge repository that grows over time
  2. Forms as a knowledge hub for the internet to use instances of my findings and observations and build cool things on top of it
  3. Making the information indispesiblity and indestructable in the coming future

But all three categories have one similarity: they all try to preserve information and channel it towards the world in a way it can stay alive for maxima. But then, there is one flaw in each points: the percieved confidence differs.

In first, I write it for myself. It is learning in progress. So the way I intake information is solely based on the learning point of view. Obviously, I would be highly confident in my own frame.

In second, I write for the people. It is channeling in progress. However, by nature, the reader assumes that not only do I know what I am presenting here, but also that I am very rightly presenting content with high confidence according to their own frame. Only to realize, at some instance, that I might make mistakes due to less experience or knowledge.

In third, I write for the future. It is preserving in progress. I want the piece to be useful2 today as well as tomorrow. So the me in the future would surely diagree with the context of the content, if I am learning better. If I don’t then that means I am not learning as much as I should. Probably, me and other people would need a metric to realise how much to believe the words I present here.

In summay though, its more like how much would you like to trust the work I present here. Higher the value, higher the trust level. Lower the value, lower the trust level. Null value means there is no use of the metric trust in that specific article (applicable for artifacts such as sitemaps, changelogs, and stuff like that..)

Footnotes

  1. which, turns out, he “stole” from Muflax’s “epistemic state”

  2. though, a better word would be relevant, because relevance is a hugely subjective and time-independent phenomenon