2 ms·
I'm not a fan of that quote. It's actually quite common for the gate's purpose to have been forgotten or for the gate to be some pathological emergent side effe
by api 10y ago
I'm not a fan of that quote. It's actually quite common for the gate's purpose to have been forgotten or for the gate to be some pathological emergent side effect of something else. Not everything is necessary or rational. In old systems there is often a huge amount of "cruft" that simply does not need to be there.
Edit: I do believe trying to explain the reason for a thing is a worthwhile exercise, but failure to do so does not prove a thing's value.
- Natanael_L 10y agoMy way of expressing the same thing would be something like this (I'm not sure if it is an improvement, though): Old methods are not automatically bad because you don't understand them or why they are used. New methods are not automatically good. If you want to change something, try to prove both that your change isn't worse in any important way and that is notably better in some meaningful way. Understanding where the original method came from helps you prove the right things.
- baldeagle 10y agoAnd that is the root of the article's thesis: the current regulator has basically said that if we had technology X back in 2007, none of this would have happened; this idea is a dangerous presumption against the adaptability of the system based in partial models.