I came across a post this weekend about different types of knowledge It’s a well written post that calls out one of my favorite topics – you don’t know what you don’t know. I think the Five Orders of Ignorance (or my tribute to the five orders here) are brilliant, and the author nails Armour’s zero, first, and second level ignorance with his take on knowledge (described as “shit you know, shit you don’t know, and shit you don’t know you don’t know). 20I (read two-oh-eye) – what you don’t know you don’t know (aka the unknown unknowns) are a big reason why we suck at scheduling. Think about it – when we throw out estimates for how long a particular task will take, it’s often half guess, and half padding because we know “stuff” will come up (yes, there’s a lot we can do to make scheduling more accurate too, but that’s a topic for another day).
IMO, there are two pretty big things missing from this article. The first is a discussion of suitable methods to determine what we don’t know we don’t know (aka Armour’s 30I). A passion for learning, critical thinking (and the knowledge or acknowledgement of 20I) all help us identify and solve the unknown unknowns. The way we learn is to discover what we don’t know we don’t know, make it something we (just) don’t know, then make it something we know. We move knowledge down through the orders of ignorance – without a method for discovering what we don’t know we don’t know, it’s just a problem that we’ll never solve (or never know about). For some, I suppose that’s fine, but not for people who are interested in learning.
The second point is that the author doesn’t acknowledge Armour. What’s great about this is that I don’t think it was on purpose – I’m pretty sure the author just didn’t know that Armour had already discussed the same topic in depth. The unknown unknown’s will get you every time!
regarding ” it’s often half guess, and half padding” — yet, most project estimates need to be to 5 or 10% acceptance level. However, with the right tools and experience, one may learn to estimate as appropriate to that level. Often, half padding is not a good idea which perhaps may mislead for future estimation exercise. However, isn’t it ideal to estimate to closest acceptance level to the best of knowledge and learn the risk mitigation approach for assumptions, possible unknowns and pad accordingly? This comes challenging for most date driven projects where date is not flexible in the tripple contraints.
regarding “second point is that the author doesn’t acknowledge Armour” — perhaps the 5OI – in the context of being ignorant of this topic was discussed before. I have read about about a similar order of ignorance in the context of feedback mechanism – where learning is still a key attribute
thank you for the link levels of Ignorance article.