4 ms·
One theme in the article is that regulated, licensed engineers don't produce buggy designs. Of course they do. Deepwater Horizon, Fukishima, to name a couple. E
by skookumchuck 9y ago
One theme in the article is that regulated, licensed engineers don't produce buggy designs. Of course they do. Deepwater Horizon, Fukishima, to name a couple. Engineers produce buggy, bad designs everywhere you look.
The reason bridges don't (usually) fall down is because most of them are copies of other successful designs, and because they are way overbuilt.
I seem to recall a bridge collapsed just a couple months ago squashing cars and killing people. It was a new design.
- bostik 9y agoI am unsure about Fukushima being a buggy design. To me it has sounded more like mismanagement and negligence than anything else. Case in point - the plant disaster happened because the tsunami was higher than what the sea walls had been designed for. From what I have learned, Japan has plenty of evidence that magnitude-9 earthquakes and the following tsunamis have happened in the past. The protections in Fukushima (and Daichii) were designed against less. To fail under expected peak load is bad engineering. To ignore evidence and go with insufficient safety margins is something worse.
- skookumchuck 9y agoThere were numerous design errors that enabled the flooding to destroy the plant. For example, venting the hydrogen gas into an enclosed area, which resulted in an inevitable explosion. There's plenty more, if you read a step-by-step account of the disaster. Engineers iterate their designs based on past failures just like software engineers do. It's just that the life cycle is so much shorter for software, the iterations come thick and fast.
- bostik 9y agoThank you, those details definitely make a case of consistently bad engineering.
- iliaznk 9y agoAnd how about cars that get regularly called back over various hardware faults?