So much blood this week. Fortunately, all of it was removed from me under medical supervision. On Tuesday, I had a bit taken for blood tests. I have been feeling a bit run down, and was wondering if my iron levels were falling. Turned out they were fine, but my glucose levels are a bit elevated – not to the diabetes stage, but heading in the wrong direction, so I will have to get more diet and exercise. Yay.
Then on Wednesday, more blood, this time donating. I love giving blood. It's such a tiny thing to do but such a wonderful little symbol of your commitment to other people, for no reward other than an orange Club biscuit at the end.
Oh, and I finally deleted my Substack. Everyone who was subscribed to it should have been ported over to WordPress. I am, though, considering whether I should move to Ghost. Because hey, who doesn't love a bit of tech-related shenanigans?
This morning we went to see Poor Things, at a 10am showing (which feels almost naughty). If you haven't been, you should go: it's the most brilliant film I've seen in a long time, with fantastic directing and performances. I could spend a couple of days trying to unpick all the threads from it, and it still wouldn't scratch the surface. Plus, Emma Stone should be a shoo-in for the Oscar.
The three things which most caught my attention
- If you're not reading Rachel Coldicutt's occasional newsletter, you should be.
- I have many feelings about the work from home vs return to office wars, but the biggest one is this: it's not a war.
- This is from last year, but basically Andrew Ridgeley is a lovely man, and what a loss to music George Michael was.
Things I have been writing
One of the biggest concerns I have about the current AI-mania is the lack of understanding of what a major change it is. Now that Microsoft has started to roll out Copilot for Microsoft 365 to all sizes of business, it's likely that more and more will turn it on (at $30 a licence) and think that's their "AI stuff" sorted. And then, of course, lay off 10% of the workforce because what their spreadsheet reckons is the "efficiency gain".
That, of course, is bunk. Using AI in your business is about people, and how you train them, and it demands a change from a "one and done" training approach to a continuous structure learning system – something that's not easy.
I also wrote a related piece about using the ADKAR change management framework to roll out AI. The point that I wanted to get across was that you are committing to a major strategic change, and you need to do that formally – and manage it, rather than just imposing it on teams. ADKAR is great for that, and, as I note, if you do it well it's not actually going to be cheap because you need potentially new roles to implement ongoing optimisation of the way you work with AI. Interesting times: it reminds me of the early days of the web.
Things I have been reading
I finished off Neal Asher's Jenny Trapdoor, which was… OK. I spotted the end coming about 20 pages in, which in a novella is always possible but a bit disappointing anyway.
Then I started and rapidly finished Stephen Baxter's Creation Node, which was very Baxter, with all that entails. I felt in places like it was a cosmology lecture masquerading as a novel, there were aspects of it which made no sense at all from a plot perspective, and had I been editing it I would have wanted some of it to just get dropped. It felt, at the end, like Baxter had created an interesting backdrop for a story but not really put much of a story into it. Which, as I said, is very Baxter.
