Hacker News Reader

Comment by PaulHoule | Hacker News Reader

One problem is that people are too macho to admit they have cognitive limits.

I know people in the finance business who are sending their CTO's out to take the Series 7 exam because to write code for some domain you need to understand something about the domain. That is on top computer science principles, your programming language, the "standard library" for your programming language, libraries you use on top of that, your build system, your version control system, and all kinds of stuff.

And now people think they are going to be left behind if they don't learn Angular, Rust, Shiny7 and all of that.

Every bit of bullshit steals your cognitive capacity and turns you into an 0.1x programmer.

View on HN

As an artist as well as a programmer, I've learned a lot about the techniques of professional artists. Most are specialists, not generalists. They have a specific, focused toolbox, and they do not leave it. If a guitarist gets that sound from this guitar and amp and pedals and scales and rhythms, he sticks to that. If a painter prefers these subjects and those brushes and paints and canvas sizes, she sticks to that.

Become proficient in a medium to the point where absolutely no conscious thought is required to use it to its fullest extent. That's the secret of great artists.

Apply that to programming, now. You can learn a field and the core tech of the field, and get a lot done for a long time. Or you can go chasing the new and shiny every six months and drive yourself up a wall and spend all your time learning instead of making.

Some short advice from an older programmer: this is the best way to ensure that you are unemployed in 20 years. Art is old (even older than me!). You can specialize in something and have a pretty darn good feeling of assurance that it isn't going to go away before you die.

Programming is a young field. It's changing dramatically from year to year. My first paying job was writing FORTH code. At one point I was writing OO C code on an embedded platform because we didn't have a C++ compiler. Eventually I became an expert in C++ (like an actual expert). A few years ago I went to a job interview and they asked me questions about Boost. I had to tell them that back when I was an expert in C++, the STL hadn't even been written. We were all using RougueWave. I've been paid to work in Perl, Python, Java, C#, Ruby, Go, Javascript and several other languages you will never have heard of because they were proprietary languages written by the companies I work for. I've built embedded systems, real time systems, distributed systems and now I'm trying to fix 10 year old Ruby on Rails code.

The thing is that none of the stuff I do is really all that different. It's just the details that change. Every time I learn something that is truly different, I always find a way to apply it to what I'm doing (if you look above you will notice a huge hole where functional programming should be... it's getting filled).

This field is about learning every day. If you want to be more than average and you want to stay that way for more than 10 years, then you have to accept that you will be spending all of your time learning. And making.

HTH!

I am an older programmer, and what you described is my career, more or less. And I fight off recruiters with a stick, switch jobs every year, and laugh at the articles about how there's ageism in the industry.

The problem I typically see isn't that someone focused on a technology for too long and became unemployable. It's that they focused on an employer for too long and became unemployable. I started my career doing C and sql 20+ years ago. I could still easily make a living at them if I wanted. My second major technology was enterprise Java. I could still make a living at that if I wanted (and to an extent, I still do).

But people spent 10-20 years in a culture. They don't know how to even look for a job anymore. Their skills may be great, but they've never put in a job application online. They're unknown to the recruiters. They don't know how to write a good resume that reflects their abilities. I watched my wife go through it last year. Laid off after 15 years at one company, her problems were twofold. First, she had narrow but deep domain expertise. Second, she was rusty at job hunting and out of touch with industry standards that had evolved without her. She still believed in waterfall and thought agile was a joke - not good for someone looking for product owner roles! She learned, but it took a while.

I don't think that those two options are mutually exclusive, you can specialize and keep up with the times...I think you kinda need to recognize the direction the times are moving in to keep up though.

I can see validity to both of your perspectives. How about the t shaped engineer concept? I like this description of the t shaped engineer: http://business.stackoverflow.com/blog/should-you-hire-t-sha...

Yes. I can agree with that. Career strategy is tricky in this business so it's best to keep a flexible mind.

Seems to me that most technical people in this business have no career strategy. They settle into a job and inertia takes hold, and a body at rest tends to stay at rest until a force is applied to move it. So they will sit their butt in a seat and be underutilized, and perform mind numbing work, with a nagging voice in their head telling them that their tech muscles are experiencing atrophy, but yet still they stay put until the bitter end.

Really applies to all people, not just tech people.

What if you make your core competence as a programmer "gaining steady footing in new tech", along with basic like code structuring, version control and some automation (CI/testing etc.)?

That way you might not become "the greatest X86 assembly programmer" or whatever niche you like, but you might become a very proficient system integrator, which is useful in its own right.

Oh, that's me. I'm a mile wide and an inch deep. So I can do a lot of things, most of them not very well. My niche friends are much better than I am at anything I can do, but I can do a lot more things.

It's worked out kind of weird, pushing me (after a long career) into being a founder, and building a tool for generalizing common problems in systems integration. :) My desire to wear lots of hats finally works! But really? I'm more or less a specialist at being a generalist. And I've learned not to chase shiny new tools, because they're very distracting.

I'm the same way! I try to get good at something, and then immediately move on. I am the best at nothing, and better than almost no one. But I am the only person I know that can write code, swordfight, surf, use a ham radio, play a handful of instruments, build things out of electronics, draw and paint, shoot, blacksmith, and blow glass (all at the same time -- just kidding)! I make myself very happy by being able to try something and then practicing until I get "good" in my own opinion. It's absolutely true that up against pros I am pretty terrible! I have also come to understand over the years that this is a byproduct of having a hacker mentality. I am loathe to use that word in reference to anything I am or anything I do, but that's just what I see. Hackers are curious creatures, and so it follows that they can do a lot of things.

It's often really hard to judge how skilled you are at something. https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect is overblown, but being able to say do a crappy weld in 4 hours that a competent person can do better in 15 minutes is often just dangerous. Many a handyman has caused 20,000+$ worth of damages or gotten someone injured.

That said, hobbies don't need to be productive and half assing something is often good enough. Yes, a simple wedge doorstop may damage the door, but well sometimes that's just not an issue.

Did you mean to reply to my post?

This is the direction I try to take. Every time I hear about a new technology I go and read the basics of it and try to understand where I might use it. Then when an opportunity comes up, I investigate and see if it's truly applicable and worth learning. In my experience, trying to dive right in to something you don't have a relatively immediate use for tends to produce a lot of wasted time and make learning slower and harder given the lack of direction.

My suspicion is that the answer is "learning isn't doing".

These are two different functional modes. The mastery of acting without conscious thought is very hard to attain. See also flow (Cziczentmihaly) and related topics.

Learning incrementally can help, and I'd structured my own career such that I was doing that. I ultimately opted out of it when that stopped being an option. I was simply spending all my time catching up, and not finding myself (nor, to a very large extent, any of my peers) actually proficient with the New Hawtness.

I've been in the industry 20+ years. The only technologies I've used consistently are Unix, and relational databases. The languages I work with, the frameworks, all of that stuff is new since I started, long ago. So yes, you have to keep learning.

But that doesn't mean you have to live on the bleeding edge, either! A tech that is proven out, used widely and stable, that's worth investing time to learn. In Crossing the Chasm terms (great book on marketing!), you need to learn early majority tech, not early adopter tech. You can be an early adopter for kicks, of course, but don't pretend it's to be more valuable.

Fair point, though many of the tools I've used tended toward evolutionary development. Came a point a bunch of stuff was being binned and/or wall-flung pasta-style.

Maybe you just aren't very good at picking languages? No offense but I have been using C, C++, C#, Python, and JavaScript for what - almost 15-20 years now?

More I hop jobs a lot. I was using C at the beginning of my career. Lately, I'm using C again (and more importantly, using arcane C skills that no one has anymore, doing a 32-64 bit port). I picked up Java in 2001. I've done a fair bit of Ruby, too. Doing Python in the current gig, which is mostly new to me but trivial to learn.

The point is, I haven't had to deal with C for a long time. It's odd to be doing it again.

The problem is you get stuck as a Junior developer in skill set if not job role. We regularly get collage interns who can often get productive in new technology in around 2 weeks of effort just like senior developers. It's the curse of specialization that the more people who can and will do something the less that thing pays.

If you were a college professor, you would generally spend most of your time reading, researching, and _synthesizing_ learning into digestible information for others. Your ability to do new and groundbreaking things would be based partially on the mastery of a discipline and partially on the ability to be constantly exposed to and challenged by new ideas.

I don't think programmers are more like professors than artists, or vice versa. These are just two different modalities applicable within the large, varied landscape that is computer engineering.

Leonardo, Michelangelo, Picasso, Mozart, Cezanne, Bach, Rembrandt, Goya - none of these artists stuck to one instrument or medium, most were masters of many and all experimented freely. Your thesis does not hold true for many great artists I can think of.

True, but then they are the all-time greats for a reason.

Some people that grew up being told that they were smart tend to think of themselves as that kind of master polymath type and end up ignoring that detail.

Most successful artists in my generation are media-agnostic and work fluidly between sculpture, installation, image-making, digital, video, and sound as media.

> One problem is that people are too macho to admit they have cognitive limits.

Yes, I experienced this and it seems to be a real problem, for various reasons: if it does not look complicated, investors won't like it, won't buy it; if it is too simple to use, people will think I am dumb. If the setup is too simple, people will think I am incompetent, etc.

Simplicity is a prerequisite for reliability, not a nice-to-have.

Well, I AM dumb. I am powerless in the face of my own stupidity.

Everything I do hinges on the acknowledgement of that fact. And you know what? My stuff works, because I can whack all the complexity down into a manageable ...widget that enforces all the rules for me. I know how to develop enough introspection of such things to keep from having real problems. I'm down to reported bugs where like the file system crashed.

No, I think too many people are still trying to win the science fair.

When my son turned 4 or 5 he got obsessed with Rube Goldberg machines which really made me sick to my stomach because at the time I spent all day debugging problems in Rube Goldberg machines.

Maybe for some people, building Rube Goldberg machines is an essential prereq to being able to code without building them.

Is it a coincidence that both of you almost quoted C. Moore?

"There seems to be a propensity to make things complicated. People love to make them complicated. You can see that in television programming you can see it in products in the marketplace, you can see it in internet web sites. Simply presenting the information is not enough, you have got to make it engaging. I think there is perhaps an optimal level of complexity that the brain is designed to handle. If you make it to simple people are bored if you make it too complicated they are lost"

http://www.ultratechnology.com/1xforth.htm

Is it a coincidence that both of you almost quoted Edsger Wybe Dijkstra?

"Simplicity is a great virtue but it requires hard work to achieve it and education to appreciate it. And to make matters worse: complexity sells better. " (EWD 896)

> And now people think they are going to be left behind if they don't learn Angular, Rust, Shiny7 and all of that.

The question you always need to ask is: "Why is <this> better than what I have now?" If you can't answer that DON'T LEARN IT.

10% better isn't enough to take up brain space. Your proficiency with your current tools easily exceeds that. 100% better probably is good enough.

The only thing I've seen in the last 5 years that has been good enough to make me start learning it is Rust.

Looking at how nice Glium made programming OpenGL was the final push. Programming OpenGL is painful on a good day, and Glium managed make a lot of the pain go away. The fact that nobody manged to do this in any other language caught my attention.

Maybe Rust isn't actually required, but I'm willing to learn something just to listen to a community/ecosystem that can produce something so nice.

Being left behind is a very real problem if you're a contractor. You just have to learn the tools that everyone is using, including Angular, React, React Native, and Node.js. There's still Rails jobs out there, but they seem to be disappearing.

> And now people think they are going to be left behind if they don't learn Angular, Rust, Shiny7 and all of that.

I have a bit of ADD-ness so shiny objects look the best. I love learning and have learned to curb my learning appetite to skills / areas that I think might benefit me in the near term.

> Every bit of bullshit steals your cognitive capacity and turns you into an 0.1x programmer.

Agree3

Reply on news.ycombinator.com