Suppose I Wanted to Kill a Lot of Pilots (2020)
newsletter.butwhatfor.com
Suppose I Wanted to Kill a Lot of Pilots (2020)
1–10 of 130 posts
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#2A lot of "security mindset" comes from this principle as well. In theory, you should have strong barriers everywhere, but we're not to that level of maturity yet. (I still hope to see some successful implementation of "capabilities" like in E or something before my career is over. That's an example of what I'm thinking of as maturity.) In practice you get a long way just by writing some code down, and then thinking How would I break this? Oh, hey, if I put this combination of characters into my decoding routine I'll get up to the SQL command level, oh, if I just assume that a user with no email is an admin and I let users change their email, they can become admin, etc.
It isn't theoretically ideal, but it's a lot better than not thinking this way.
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#3Good advice. Looking at things from the opposite perspective helps a lot.
It’s easier to see where things go wrong than where they go right. You can just list wrong thing after wrong thing and a solution appears once you’ve scoped out the things that don’t work.
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#4A lot of "security mindset" comes from this principle as well. In theory, you should have strong barriers everywhere, but we're not to that level of maturity yet. (I still hope to see some successful implementation of "capabilities" like in E or something before my career is over. That's an example of what I'm thinking of as maturity.) In practice you get a long way just by writing some code down, and then thinking H…
Yes it's called being adversarial and it's the best way to improve something.. by actively looking how to break it
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#5Previously: https://news.ycombinator.com/item?id=27720912
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#6TRIZ practitioners quickly realized that the root cause of some problems cannot easily be identified, which makes a frontal assault on them impossible. This kind of problem is solved with TRIZ’s Anticipatory Failure Determination mental model — rather than solving for success, solve for failure.
This is pretty darn close to https://en.wikipedia.org/wiki/Apophatic_theology : I can't tell you what The Good Thing is, but I can tell you many things that it is _not_.
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#7Pre-mortem exercises (opposite to the typical post-mortem, after something has gone wrong) can be a really fun process of discovery and ideation.
Unfortunately though, I have yet to run one that actually helped us avoid the exact risks we discovered!
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#8A lot of "security mindset" comes from this principle as well. In theory, you should have strong barriers everywhere, but we're not to that level of maturity yet. (I still hope to see some successful implementation of "capabilities" like in E or something before my career is over. That's an example of what I'm thinking of as maturity.) In practice you get a long way just by writing some code down, and then thinking H…
In my experience, TDD has a bit of a similar effect on your perception on a problem.
By first designing the test that would make the code fail, then making sure it fails, then create the code to have the test pass, you kind of change the way you think about the original problem, and it sometimes helps.
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#9Good advice. Looking at things from the opposite perspective helps a lot. It’s easier to see where things go wrong than where they go right. You can just list wrong thing after wrong thing and a solution appears once you’ve scoped out the things that don’t work.
Totally agree. Charlie Munger studied the big ideas of several disciplines, and then applied it into his daily life. He applied the power of solving problems by inversion, which is an idea commonly used in mathematics, to the situations he encountered in real life. As he says "Invert, always invert".
Re: Suppose I Wanted to Kill a Lot of Pilots (2020)
#10Pre-mortem exercises (opposite to the typical post-mortem, after something has gone wrong) can be a really fun process of discovery and ideation. Unfortunately though, I have yet to run one that actually helped us avoid the exact risks we discovered!
The goal of the plan is the planning. Pre-mortems train you to think of what could go wrong to build general problem-solving strategies and tooling. The key is to not over-index on any one particular possible outocome or type of scenario.
From TFA:
> Now you have a list of actions to avoid. Avoiding them begins to move you away from failure and closer to reaching your goal.
Alternatively: Now you have a list of bad things and some ideas of what you would do or tooling you would need when bad things happen.
In the good case this is "boy scouts being prepared." In the bad case this is "hoarder preppers detached from reality with unnecessary firepower."