33 ms·

This is shortsighted. The reason deep knowledge is more valuable is that learning anything stops being fun, and therefore presents less of a biological pull, af

This is shortsighted. The reason deep knowledge is more valuable is that learning anything stops being fun, and therefore presents less of a biological pull, after the first few milestones. There's a logarithmic decay in reward to effort put in. So naturally you're going to want engineers that have put in the hard work to build their knowledge, rather than the posers who read about engineering but never actually got into the nitty gritty. Why? Because it becomes rare, and scarcity drives up value. Simple! EXAMPLE: Why would I care if my plumber also moonlights as a surgeon if all I care about is getting my faucet fixed without him passing the buck due to perceived complexity? Those things are great in a dinner party setting, but in the professional workplace you want professionals not dabblers. Whenever I meet a self styled "generalist" nowadays I scoff. It's not hard to read HN daily and pick up on the lingo and memes and fake your way to 99% of a "generalist" in both knowledge and impression, but that doesn't make one a good engineer.

kelnos 6y ago
I think your premise is flawed. Generalists are not "dabblers". They are professionals just as much as specialists are. Either you have a very odd (and incorrect) view of how software development is done, or you're presenting a straw man that doesn't have anything to do with the article's premise. I find most specialists useless when it comes to architecting and building a full system, and often when they are a part of a team, focused on their specialty, they build in a way that is difficult to fit into the whole. Obviously there are good things, too: my knowledge of things like data science and graphics is limited; if a part of a project called for deep knowledge in those areas, I'd be happy to have a specialist on the team. But unless they are more of the "T-shaped" type the article talks about, they need a lot of hand-holding when it comes to anything outside their domain. Generalists bridge that gap, and handle necessary concerns that most specialists don't even consider, let alone know how to handle. (Full disclosure: I consider myself a generalist, though at times I have ended up diving deep into particular areas when projects have required it.)
nucleardog 6y ago
You’re not arguing against a generalist (varying depth over a handful or many areas), you’re describing, sure, let’s call it a poser or dabbler (little to no depth in any area).

kelnos 6y ago

I think your premise is flawed. Generalists are not "dabblers". They are professionals just as much as specialists are. Either you have a very odd (and incorrect) view of how software development is done, or you're presenting a straw man that doesn't have anything to do with the article's premise. I find most specialists useless when it comes to architecting and building a full system, and often when they are a part of a team, focused on their specialty, they build in a way that is difficult to fit into the whole. Obviously there are good things, too: my knowledge of things like data science and graphics is limited; if a part of a project called for deep knowledge in those areas, I'd be happy to have a specialist on the team. But unless they are more of the "T-shaped" type the article talks about, they need a lot of hand-holding when it comes to anything outside their domain. Generalists bridge that gap, and handle necessary concerns that most specialists don't even consider, let alone know how to handle. (Full disclosure: I consider myself a generalist, though at times I have ended up diving deep into particular areas when projects have required it.)

nucleardog 6y ago

You’re not arguing against a generalist (varying depth over a handful or many areas), you’re describing, sure, let’s call it a poser or dabbler (little to no depth in any area).