A vivid memory I have from growing up in the 1970s was my father’s what seemed like annual slide deck preparation.
This is, of course, in the days before Harvard Graphics, let alone PowerPoint.
Dad was a lecturer at the University of London. He had to prepare lectures for his MSc students, and prepared visual aid slides to go alongside his lectures. The hard work that I didn’t see was the preparation of the lecture material itself, but the visible part was making the slide deck.
It was a laborious process. Dad started with sheets of white paper, and packs of Letraset. If you are not old enough to remember Letraset, it’s very much in the category of “stuff that I’m not going to bother trying to explain to my children” alongside phone boxes, fax machines and the fact that TV used to stop in the middle of the night when there were only three channels to watch.
After creating his slide layouts, he’d then photograph them, using a monochromatic film that would take pictures (in negative) in either black or white (no grey). This would then get sent off to the developers, and back would come 35mm transparencies that had white text on a black background. The final stage would see dad using tiny bits of coloured gel to highlight particular words in particular colours, and the bits of film being put into 35mm slide cases, which in turn would be loaded up in order into a carousel which could be taken into the university.
The slides could then be presented in a lecture room, presumably with relatively low lighting, as dad gave his lectures.
What a palaver.
However, with such an effort involved, what I can be fairly sure of is that dad’s preparation that I didn’t see, thinking about the structure and content of the presentation to which slides would be applied, was pretty darn meticulous. It had to be because mistakes would be costly.
No hours were spent arsing about with fonts, colours and themes back in those days.
I’ve been thinking about this huge effort with news coming out of all of the whizzy AI that Microsoft are going to be applying to PowerPoint (amongst other tools) in what they are calling “Copilot“.
As they put it, “Copilot in PowerPoint helps you create beautiful presentations with a simple prompt, adding relevant content from a document you made last week or last year”.
I’m not sure, though, that the world needs more “beautiful” PowerPoint presentations.
Over the years I’ve come to the conclusion that PowerPoint is used in two main ways. The first is the way that I think PowerPoint people think it is used – beautiful people giving beautiful presentations, winning deals, winning hearts and changing minds.
But mostly PowerPoint is used as a way to create documents that are to be consumed on screen. A PowerPoint slide has become a unit of content, as friend Chris King so eloquently puts it:
Adding AI to PowerPoint feels like throwing a solution at an ill-defined problem, a little like the way Microsoft made Paint into a 3D tool a few years ago because the future was going to be 3D if they were going to sell any HoloLens devices.
If we actually look at the two predominant uses of PowerPoint, however, I’d suggest we need different things.
To make a beautiful presentation you need to start with a beautiful story – Marcus Brown’s Presentation Canvas is a far better starting point for that need than an AI paraphrasing the terrible business case you wrote last week. If you spend your time planning what you are going to say, you are going to deliver a far more compelling message than if you leave your message to an AI. Trust me on this.
But really interestingly, the “create documents to consume on-screen” need is a seriously unsolved problem. Microsoft tried a thing called Sway, but you probably haven’t heard of that. PowerPoint has dominance but simply isn’t very good at the task. And Word still formats things for a world of print.
The simulated intelligence tools that are emerging at the moment are really fascinating and very clever. But they are solutions looking for problems. My hunch is that it will be a fair while before we see truly useful implementations, and my other hunch is that they’ll only get useful when we stop trying to get them to do things that humans are better at doing. Like storytelling.
mmitII 
One thought on “Slide decks”