When I was about to start a task, I was proactively distracted by a really interesting and topical YouTube recommendation.
By way of context, in week 20 of TM354 Software Engineering (which is an optional bit of study materials) students are encouraged to listen to a talk by Grady Booch about the topic of software architecture. The recommendation was to listen to a talk from The Pragmatic Engineer by Grady entitled The third golden age of software engineering – thanks to AI.
It is worth a listen. Booch is known for his work on UML, which features heavily in TM354. Over the last couple of years (before I left X), I’ve been aware of Booch’s opinions about Generative AI. I’ve shared his cynicism about the hype that has been shared by the developers of GenAI systems.
It is worth looking at the bookmarked sections of the video. Each of his descriptions of the different ‘golden ages’ are interesting. In it, he highlights drivers and change. In fact, I really like how he introduces the origins of software engineering.
In his talk, he responds to an assertion from the CEO of Anthropic who suggests that AI can take over tasks that are carried out by software engineers in 12 months. Given that people (namely, software engineers) are able to know and understand the richness of the human contexts in which software exists, and be aware of the nuances that can be expressed within sets of non-functional requirements, I hold the view that this is nonsense. A general purpose generative AI tool cannot provide insight after speaking with a number of different stakeholders, and appreciate how begin to make changes to existing systems. Booch presents his response to Amodei’s assertion by adopting more scientific language, which is certainly worth listening to.
Some of the themes that run through his talk (in addition to AI) are that of abstraction, ethics, and change. He also offers some helpful suggestions about ‘new skills engineers will need to succeed’.
If you’re a software engineering student, and have a free hour (which might be easier said than done), Booch’s talk is worth a listen. Some bits can get pretty dense in terms of detail, but don’t worry out that too much. Sometimes it is useful to listen for the gist, since it’s always possible to go over it again later.
As an aside, he also mentioned a website called Victorian Engineering Connections. Following his comment, I searched for George Boole, and watched a web of links appear on my screen.
To conclude, here’s a quote from Booch’s X account: “The story of computing is the story of humanity: this is a story of ambition, invention, creativity, vision, avarice, power, and serendipity, powered by a refusal to accept the limits of our bodies and our minds.” I really like that. After all, at the end of the day, software engineering is all about creating products and systems for people, by people.