Lets put this in terms of web development: You wouldn't start building a webapp by building your own web framework first.
Umm, yeah, I kinda did this. And as I am not ready to promote either the framework or the app, I can certainly relate to the OP.
Having said that, I don't regret my decision at all, and I'm now using the framework in some production work for clients. This is very satisfying.
I also have a mortgage and two kids, and I am paying the bills while my wife is in grad school. So my side projects must remain just that. But they keep me sane, so for me it's not just about the end result.
Having kids makes moonlighting much harder than having a day job does, IMHO. (But what do I know, I've been freelancing since Katrina.)
And I really am going to publish them this year! Really!
If I were starting from scratch TODAY, then I agree with you.
I started a small ecommerce site back in 2003, and django, etc didn't exist. So I did develop a framework from scratch in C. 10 years later, it's the engine my new data storage startup (Nuevo Cloud).
Point is, things like web app engines and game engines aren't things that pay off immediately. They require _a lot_ of development, and during the entire time that you're working on it, you're basically handicapped until it catches up with other engines/frameworks. But after it's far enough along, it can be a big advantage, letting you do something unique compared to your competitors.
Re: your last sentence: Isn't it even more likely that it could end up being a complete waste of time and effort and be inferior to other frameworks/engines?
The main takeaways of the article are "have a business model" and "don't build your own engine." It adds years to the process, it's extremely difficult to get right, and the end result will be worse than using existing tools.
If your goal was to make a web app, would your first step be to write a garbage collector and a database management system? Would you invent a templating language, and build a parsing engine for it in C? Would you think it responsible to pay a web developer a salary to do those things if their goal is to just make a website? If not, then it'd pretty foolish to pay yourself to make a game engine.
The decision to invent should be made intentionally, rather than impulsively. If your goal is to have fun, then have at it. But if your goal is to launch a product, it's not prudent.
I've had a lifelong dream of building my own game, and I'll be sad if I die without trying. I'm way less prepared than the author. I've done web dev for 15 years. My last game was an ASCII roguelike in GW BASIC. I'm almost done reading Game Engine Architecture by Jason Gregory, and building my own engine sounds like a blast---but also like a 20 year project. Fortunately my dream game is more like an old 2-D Zelda 1 or 3. Still, it's tempting to build even that with 3-D techniques. . . .
The big lesson I'd apply to myself is to build something modest. That seems even more true given my lack of experience, although I'm not sure I'll listen.
I guess the difference between the author and myself is that I fortunately don't have the expertise to think of game making as anything other than a creative hobby.