Commoncog This Week
This week's Commoncog piece is short (and free); next week's piece will be members-only.
I'm currently travelling and recovering from the flu, so I don't have the capacity for a full update this week. I'll keep this one short.
Yesterday, I hosted a private, Commoncog members-only Q&A on Software Dark Factories.
If you're a Commoncog member and you missed this due to timezone issues, the recordings are in that post, past the paywall. You may click here to jump to them.
I just want to talk a little about what it felt like as the session was ongoing.
But first, some context.
How do you make sense of a fast moving technology?
A few weeks ago I finished a three part series on sensemaking. Sensemaking is defined as 'the deliberate attempt to understand events'. My goal was simple. We are faced with a fast moving, fast changing technology that has the potential to affect our business or our careers. It can feel overwhelming. What do we know about how to make sense of fast-moving events?
The approach that I took was (I think) common sense: I first asked "what do we know about good sensemaking?" The answer: the best theory we have about sensemaking comes from some research funded by the US Military — Gary Klein et al's Data-Frame Theory of Sensemaking. In a way this makes sense. If you're a warfighter, you want to be able to make sense of the battlefield, under conditions of extreme uncertainty. If you are the world's best funded military, you'll probably want to improve your soldiers's sensemaking capabilities.
(Sensemaking also turns out to have deep implications for accelerated expertise, but that's a story for a future essay).
The way humans make sense of war (and firefighting, and surgery, and policing) is the same way you make sense of technological changes in business.
So the sensemaking series was straightforward to write:
- First, tease an approach to make sense of AI
- Then, introduce a framework to understand how sensemaking actually works, along with how experts sensemake and how this differs from novices
- Finally, apply what we learnt in step (2) to approach (1), by way of a real world example.
The real world example that I used was software engineering. I wrote that there is an emerging frame that is particularly concerning: 'Software Dark Factories'. What this means is a software development approach where no human writes code and no human reads code.
You just ... write feature requests? Or perhaps specs? You talk to the AI? And then somehow good, correct code comes out the other end?
This is heretical and unacceptable on so many levels I don't even know where to begin.
But with the Data-Frame Theory of Sensemaking, we have better language to deal with this:
- You may construct a new frame, in order to understand what these proponents at the very frontier are doing and why they are doing it ... on their own terms.
- You may hold this frame in abeyance, without giving up your primary frame, and spend time elaborating it with new data as it emerges without necessarily accepting it.
- You may then reject that frame later, when it becomes clear that it's unworkable. Or you may accept it, replacing your current frame. Either way, you've given the new approach a good shake.
In my third and final essay in the Sensemaking Series, I wrote that I did not fully buy into the Software Dark Factory frame. I find it difficult to believe that you can produce high quality software at scale and with high velocity without any humans reading the code.
It felt too good to be true. It felt like a violation of everything I knew about software development.
And then James Cham of Bloomberg Beta offered to facilitate a Q&A with the folks who got the furthest with the Software Dark Factory approach.
What it felt like watching the Q&A and demo
That Q&A happened yesterday, Jun 1 8pm PT / Jun 2nd 11am SGT. The call was with Jay Taylor and Navan Chauhan, formerly of StrongDM. They have now left to start their own company, Software Dark Factories Inc, to help other companies build their own software dark factories.
They showed us a demo, and then they took questions.
How did the session feel?
It felt like watching someone die.
Or, more accurately, it felt like watching my idea of software engineering die.
It was very very uncomfortable.
(I was also falling sick, so maybe I was conflating the two feelings? Hmm.)
Jay addressed this a little by talking about how he had an identity crisis whilst working on this at StrongDM. I felt like I was having an identity crisis during the event.
I know I talk about how a viable strategy to dealing with new technological changes is to elaborate a second frame, that you don’t necessarily accept. I’ve written about how dangerous it is to sensemake from your prior frame when one of the anchors of your old frame may have been violated, especially in the context of a fast moving, potentially disruptive technology.
But it is still NOT comfortable to experience this in real time. And bear in mind this is in the context of a secondary frame that I haven't fully accepted!
If you are a Commoncog member and a software engineer, and you've watched the recording, I'd love to hear from you. We have some responses trickling in the forums here. One member has already written a public blog post with his reactions. Another has tweeted about it. I expect more to emerge over the next few days.
I'm still trying to figure out what I think of all this. I may have a more coherent response when I'm fully recovered.
💡 The Commoncog Membership Program is like an ongoing MBA, for a fraction of the price. Get full access to members-only articles, a rich and growing case library, plus an exclusive, members-only forum.
|
In Other News
Travelling with an eight month old turned out to be easier than expected! (Though perhaps harder than travelling with a three month old). Thanks to everyone who reached out with notes from travelling with kids. It appears things only get harder once they start walking. 😬
Alas, I fell sick on the last leg of my trip, I think from the flu, and now baby is sick also. But I think that's independent of travel woes.
If all goes well, I'll see you next week.
Warmly,
Cedric
|
