I've found his Twitter very frustrating to follow sometimes, but I also learn a ton from the things he links to and talks about. For example, I'm certainly not in the anti-OO camp, but his perspective keeps me on my toes and has been informing my programming decisions to a degree. He also seems like a good person in general — just one who happens to be an old school programmer with a very specific view on how things should be done. (As are many other game developers, I've been discovering!) People are complex and multifaceted, and you don't have to like every part of them. It's clear that he's a visionary and a fantastic game designer, and that he's been in the trenches for a while, so he's worth listening to for those reasons alone.

Anyway, it's kind of weird talking about jblow when I know he'll be reading this. Even if he doesn't like us so much anymore. :(

Yeah, I've found that programmers who work a lot in C++ are a different breed from those who tend to work with higher level and GC'ed languages more. It really boils down to the task at hand I think. There are always going to be people who want explicit control to those who want the system to figure out things for them.

Why doesn't he like us?

Every time he posts someone replies to say he doesn't know what he's talking about. (possibly not realizing who they're replying to)

I don't think he thinks most people here are "good programmers".

He's probably right. HN has some serious talent around, but it also has as rather core principles a lot of stuff that's deleterious to the kind of understanding that forms the basis of what he considers a good programmer to be.

I have throughout my career tried to maintain a grasp of everything in my stack (I went through a "study the code for HotSpot in-depth" phase that, while I don't regret it, I'm glad is over--though I may be about to start in on MRI...) and I try to make intelligent decisions about when to approach problems at different levels. I see a lot more trend-following and operating off of tribal conventional wisdom than I would like.

Well, "most" programmers are not good. So, its probably true, but for a different reason.