This was back in a time when one guy could be familiar with every single line of code running on a computer, from user space programs right down to the kernel. This level of simplicity surely did wonders for reliability. Not to say you couldn't achieve good results with a more complex modern system, though to spend a lot more time making things failsafe.

> This was back in a time when one guy could be familiar with every single line of code running on a computer, from user space programs right down to the kernel.

This is still possible, by the way. Humanity hasn't lost the ability to create simple computers. :) We just see less of them because the complex ones are more useful for day-to-day life.

I'd argue that more complex ones (appear) beneficial to programmers more than anything else (users don't appear to benefit from complexity).

Exactly what I thought. This is what real engineering looks like.

As an engineering approach to generating power, thermopiles are considered differently today than in the 1970's due to environmental and health risks associated with plutonium and the probability of a launch failure.

I agree with your assessment if you're talking about perception, not actual risk.

Shortly before Cassini was launched there were some people on my college campus protesting the mission on grounds of risk of radioactive debris spreading if the launch failed. Motivated by my distaste for fearmongering, I did some reading and found that nearly all the scholarly research and engineering studies suggested the risk was infinitesimal, and in a worst case scenario the spread of dangerous concentrations of radiation would be highly localized.

Aside: Bill Nye (himself a trained engineer) also uses his position as Executive Director of The Planetary Society to advocate for the use of nuclear power generators in deep space missions.

In light of recent events in Japan, engineering risk assessments regarding the hazards posed by industrial uses of plutonium are not exactly gospel. The fact that 3 of 4 failures of the Titan IV rocket occurred following Cassini's launch, the prelaunch engineering risk assessment for Cassini is likely to have underestimated it's risk in a similar manner.

[http://en.wikipedia.org/wiki/List_of_Titan_launches]

Cassini was launched by the IV-B variant of the Titan. Given that only one IV-B launch had taken place prior to Cassini, I'm guessing they applied a pretty big risk factor to the launch vehicle.