[TechTalk] Why Technical Writing Matters More in the Age of AI - EP083

So we're talking about the scare skill of technical writing.

You may speak English.

You may converse very well in English, but it's very, very different if you're not able to express yourself technically in English or in any language for that matter, right? You can take German, you know, it's very difficult or it's very different speaking German.

as opposed to writing and expressing yourself accurately in German.

And so this is what we're talking about.

And we talk about the contrast of how AI is now becoming a de facto gap filler for potentially our lack of or not required skill of having to express ourselves on paper technically.

But I have a very, and I don't believe that's the case.

I believe that there are many, many reasons and instances where that is actually not even something that can be good for you as a technical professional and even for your self, you know, image in terms of vis-a-vis your colleagues.

that is not a good look if you're not able to express yourself technically on a piece of paper so it it is definitely something that we have to watch out for and we're gonna kind of go through an anchor story because i do think this was very very starkly obvious that technical skills technical writing is still a big deal so The year was September 1999, and NASA's Mars Climate Orbiter arrives at Mars after traveling hundreds of millions of kilometers.

Then, all of a sudden, NASA loses signal with the rover, or the orbiter.

And then this, you know, finger-pointing starts, and, you know, what happened shouldn't have happened, and so forth, and...

But then something obvious comes up is that there was one part of the system that was working in pound force seconds, you know, in terms of energy.

as opposed to Newton seconds.

One is imperial, the other one is not.

And the spacecraft approached Mars essentially at the wrong trajectory and was lost.

And NASA's investigation identified the failure due to using metric units in the ground software being the root cause as opposed to imperial units on the actual orbiter itself.

And of course, if you're an engineer or you're a technical professional, you know that change of units, specifically from American to, excuse me, to non-American units, is a big difference, right? And, you know, pound seconds to Newton seconds is not a really good mistake to make.

And this is where the story kind of hits home for a lot of us as engineers.

I do remember another story where radians to degrees was a really big thing because the conversions was happening throughout the software.

And instead of appending radians, the parameter name, underscore radians, underscore degrees, in a way to cross-check what was happening, you had conversions all over and you really lost touch of what was actually happening in the code.

And there was a point where things were not working at flight test.

And this engineer screamed the bottom, at the loudest scream ever, he's like, what is the unit? What is the unit? What? Is it pounds? Is it radians? Or is it degrees? And we all, I mean, he totally lost it.

I mean, I think he had a mental breakdown due to the pressure and the fact that, you know, unmanageable code because of lack of technical writing was essentially affecting someone's well-being, mental well-being.

So there is definitely something to really talk about here because we are, of course, we're talking about specifically NASA and what they are showing.

There is specifically this diagram whereby it looks at what What are the different interface design descriptions? So an IDD, an interface design description, is essentially a way of communicating the interface between different very important components or very important subsystems or systems, and in so doing ensure that the development of the one is constantly related to the development of the other.