Hacker News new | past | comments | ask | show | jobs | submit login

As a kid, mainlining Heinlein, I just assumed we'd have a moonbase by now and that it would be up to something important and useful. In my 20s, I assumed that our then-primitive software engineering techniques would be refined until we could make things that were simple, polished, cheap, and reliable.

So it's a little wild to me to see software not only get more chaotic, but influence hardware as well. All in service to a creeping managerialism that runs on goals that, to the extent they can be articulated at all, get more detached from any sane human purpose.

I know shit about Artemis and would love to believe Maciej is totally wrong here. But it fits with so much of my experience of the world that it seems very believable to me.




Simplicity as a feature.

Like TFA says, if you want something to work reliably, keep it simple.

But simple isn't impressive. Tackling complex problems in complex ways is what gets recognised and rewarded. Humans are weird.


For sure. And to me that's related to the spread of managerialism and MBA thinking. One of the fundamental beliefs in that paradigm is that management is universal; an expert manager can manage anything. (This is in the contrast to the view that you need domain expertise to be effectively in charge of something.) I think this falls down because, not understanding the substance of the work, pure managers have to go by proxy indicators, like the polish of the presentations, the amount of confidence expressed, and the general wow factor.

People with a lot of engineering experience are suspicious of complex solutions to complex problems. They know the value in iteration and testing. So even if an engineer is pushing a complex solution (for resume reasons or just love of the fancy stuff), they can be reined in by senior engineers. But in the MBA mindset, a complicated solution is an opportunity to have big budgets and lots of excitement. Slow feedback loops are even better, because they can produce shiny documents, get promoted, and move on before the problems become obvious.


Agree completely. Unfortunately it seems to be impossible to build large organisations without creating the sort of incentives that feeds this kind of thinking.


> until we could make things that were simple, polished, cheap, and reliable.

That was the original Apollo mission. We went to the moon 6 times.


Apollo wasn't exactly cheap.


As the article explains, Apollo was cheaper per mission than this.


Yes. And a Rolls-Royce is cheaper than a Lamborghini.

Ie A can be cheaper than B, but still be expensive.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: