There was a time not too long ago when, once you learned how to do something, you could feel fairly confident you’d still know how to do it a few months later. I’m not talking about learning how to design a website and then sitting around and not designing anything for a decade or two and then being shocked to discover thing had changed since you last broke out your html skills. That would be understandable. But a break of a few months? How much could change in that amount of time?
In today’s technology environment, quite a bit.
As I’ve been using my iOS app to keep track of my media habits, I’ve discovered a few hiccups. The biggest one is that it has been unable to fetch information about the number of seasons and number of episodes per season for TV shows I hadn’t already started to log. Something in my code wasn’t working for new shows. (I discovered this when I tried to log The Great British Bake Off, if we’re being specific here. And at first I wondered if it somehow didn’t like British shows. It didn’t, since it also didn’t want to get data for The X-Files, either.)
No problem! I’d just sit down and work with Claude to get it fixed, right?
Wrong. Because when I went to use Claude Code via Cursor, which is how I’d taught myself before, I discovered that Claude Code and Cursor didn’t like each other anymore. It was impossible to get anything done that route unless I (you guessed it) paid more money. And that wasn’t going to happen. So I turned to Claude to figure out how to solve the problem. It took about 2 hours, but now I’ve got Claude Code running directly in a terminal window, and all is right with the world again.
Still, it was fairly jarring to suddenly discover that the entire workflow I’d developed and used for months was no longer valid. Yes, I could use AI to get around that, but it highlighted for me the importance of being flexible right now. Back in the day, you could do some research into the best platforms for just about anything, and that research would stay valid for a while. At least a few years, right? New things just didn’t come out fast enough for it to make the research obsolete.
Now, everything is changing on a monthly basis. (Sometimes on a weekly basis.) One AI platform will be all the rage in one area one day, and a month later, it’s practically quaint. It can be tempting to get (very) frustrated and give up on things altogether. What I’m trying to keep in mind instead is that now, all you have to do is wait for a month or three, and everything will change again.
I do believe things are tightening in the great AI arms race. The money isn’t quite flowing as freely as it did a year ago, and so AI companies are tightening down on what they will and won’t do. They’re doing their best to try to force people to stick with them instead of running off after the latest and greatest. I don’t know how successful they’ll be in that. I expect we’ll see some companies start to fold and consolidate, but I still don’t see anything that leads me to believe this AI stuff is a flash in the pan.
In the meantime, it was a helpful reminder to me that I need to be exercising these skills on a regular basis.